Skip to main content

DESARROLLO, OPERATIVA Y GOBERNANZA DE DATOS PARA SISTEMAS SOFTWARE BASADOS EN APRENDIZAJE AUTOMATICO

PID2020-117191RB-I00

•
Funding agency name Agencia Estatal de Investigación
Funding agency acronym AEI
Program Programa Estatal de I+D+i Orientado a los Retos de la Sociedad
Call Proyectos de I+D+i (Generación de Conocimiento y Retos Investigación)
Call year 2020
State Plan 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Beneficiary centre UNIVERSITAT POLITECNICA DE CATALUNYA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 88
Encontrada(s) 1 página(s)

UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

The knowledge graph lifecycle in NTT DATA

  • Flores Herrera, Javier de Jesús|||0000-0002-2998-9962
  • Jamin, Emmanuel
  • Nadal Francesch, Sergi|||0000-0002-8565-952X
  • Romero Moral, Óscar|||0000-0001-6350-8328
The Semantic Business Unit (SEMBU) in NTT DATA aims to increase the semantic interoper ability and accessibility of European institutions’ data projects by following Linked Open Data (LOD) principles to build controlled vocabularies and produce Knowledge Graphs (KGs). One of its most notable projects revolves around the CORDIS portal1, which publishes information about research and innovation projects funded by the European Commission. SEMBU pursues two main goals: (i) expose semantic data related to CORDIS via a SPARQL endpoint that facilitates access and reuse of quality scientific-related data, and (ii) design an efficient, incremental, and automated KG lifecycle to be used as a reference in other data projects. To that end, we have adopted state-of-the-art semantic technologies to support the creation and management of the KG with the goal of centralizing knowledge and providing an overall view of data assets that improve data governance, maintenance, and external interaction by data consumers. We have also identified some of their limitations which are tackled via an industrial PhD. This paper reports our experience, the obstacles, and proposals for generating and maintaining the CORDIS KG., This work was partly funded by the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RBI00 (DOGO4ML). Javier Flores is supported by contract 2020-DI-027 of the Industrial Doctorate Program of the Government of Catalonia and CONACYT’s scholarship. Sergi Nadal is partly supported by the Spanish Ministerio de Ciencia e Innovación, as well as the European Union - NextGenerationEU, under project FJC2020-045809-I /AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Capturing analytical intents from text

  • Pons Recasens, Gerard|||0000-0003-2225-3255
  • Dimic, Miona|||0009-0008-1869-3585
  • Bilalli, Besim|||0000-0002-0575-2389
The ability to extract valuable information from data is crucial for organizations and individuals who want to remain competitive in a constantly evolving data-driven environment. However, some of them lack the skills required to appropriately leverage the existing data analytics tools and methods. This problem is aggravated when the users are domain-experts but completely unfamiliar with data analytics terminology, as existing assistant tools, such as AutoML or Intelligent Discovery Assistants, require them to state their analytical intent (i.e., the type of data analysis they want to perform). To address this problem, we propose to capture the underlying analytical intent from textual problem descriptions by leveraging Large Language Models (LLMs). To this end, we propose a hierarchical categorization of analytical intents, along with a data collection methodology to obtain analytical problem descriptions for all of them in order to validate different approaches that aim to extract such intents from text. Next, we compare the performance of state-of-the-art approaches with LLMs, and then study the performance of different LLMs based on their characteristics and the impact of the source of validation data. Finally, we develop a prototype to showcase how our method could interact with existing AutoML systems., Gerard Pons is supported by the EU’s Horizon Programme call, under Grant Agreement No. 101093164 (ExtremeXP), and Besim Bilalli is partially supported by the DOGO4ML project, funded by the Spanish Ministerio de Ciencia i Innovación under the funding scheme PID2020-117191RB-I00/AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Evaluating quality of disparate data sources: A discord-driven approach

  • Akter, Yeasmin Ara|||0000-0002-9865-0700
  • Abelló Gamazo, Alberto|||0000-0002-3223-2186
  • Jovanovic, Petar|||0000-0003-4635-6646
  • Sagi, Tomer
  • Hose, Katja
Among other measures of data quality, determining the reliability of conflicting values from different sources is especially challenging. Traditional data fusion approaches often infer correct values in simple cases, but struggle to handle variations in data granularity (such as differences in temporal, spatial, or categorical aggregations) and offer limited insight into the nature of disagreements. Thus, we propose a new source evaluation approach for numerical attributes that measures discordance (i.e., the extent to which sources differ from each other). Unlike existing methods that focus solely on point estimation, we allow both fine-grained and coarse-grained analysis, allowing more sophisticated data quality assessments. We employ a linear programming solver that transparently adapts to any data alignment expressed in a set of operators resembling relational algebra. Extensive experiments on real-world datasets demonstrate that our method generalizes existing truth discovery techniques measuring differences with Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and can adapt to diverse and complex scenarios., Y. A. Akter is funded by the EC Horizon 2020 research and innovation programme (DEDS: grant agreement No 955895). A. Abelló and P. Jovanovic are funded by the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RB-I00/AEI/10.13039/501100011033 (DOGO4ML) and the EC Horizon Europe programme (ExtremeXP: grant agreement No 101093164)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Multi-agent debate strategies to enhance requirements engineering with large language models

  • Oriol Hilari, Marc|||0000-0003-1928-7024
  • Motger de la Encarnación, Joaquim|||0000-0002-4896-7515
  • Marco Gómez, Jordi|||0000-0002-0078-7929
  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
Context: Large Language Model (LLM) agents are becoming widely used for various Requirements Engineering (RE) tasks. Research on improving their accuracy mainly focuses on prompt engineering, model fine-tuning, and retrieval augmented generation. However, these methods often treat models as isolated black boxes - relying on single-pass outputs without iterative refinement or collaboration, limiting robustness and adaptability. Objective: We propose that, just as human debates enhance accuracy and reduce bias in RE tasks by incorporating diverse perspectives, different LLM agents debating and collaborating may achieve similar improvements. Our goal is to investigate whether Multi-Agent Debate (MAD) strategies can enhance RE performance. Method: We conducted a systematic study of existing MAD strategies across various domains to identify their key characteristics. To assess their applicability in RE, we implemented and tested a preliminary MAD-based framework for RE classification. Results: Our study identified and categorized several MAD strategies, leading to a taxonomy outlining their core attributes. Our preliminary evaluation demonstrated the feasibility of applying MAD to RE classification. Conclusions: MAD presents a promising approach for improving LLM accuracy in RE tasks. This study provides a foundational understanding of MAD strategies, offering insights for future research and refinements in RE applications., This work has been supported by funding from the HIVEMIND project – Horizon Europe call HORIZON-CL4-2024- DIGITAL-EMERGING-01 under Grant Agreement Number 101189745; and by the Spanish Ministerio de Ciencia e Innovación under project funding scheme PID2020-117191RB-I00/ AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

RE-Miner: Mining mobile user reviews with feature extraction and emotion classification

  • Motger de la Encarnación, Joaquim|||0000-0002-4896-7515
  • Tiessler Aguirre, Max
  • Oriol Hilari, Marc|||0000-0003-1928-7024
  • Bertolín Rico, Irene
In the context of app stores, user reviews are pivotal on supporting multiple requirements engineering tasks. Among these, feature extraction and emotion classification play a crucial role in requirements prioritization, feedback gathering and release planning. Empirical evaluation of these techniques is impeded by data collection complexities and a lack of reproducible methods and available tools. Furthermore, existing studies often focus on isolated tasks, hindering a comprehensive analysis of user perceptions. This paper introduces RE-Miner, a work-in-progress tool integrating multiple feature extraction and emotion classification innovative methods, enabling a detailed analysis of feature-oriented user feedback. RE-Miner comprises a web-based service for task integration and comparison, and a web application for persistent storage and analytical visualization of reviews. As a result, RE-Miner provides a platform for seamless integration, replication, and comparison of review mining techniques, fostering advancements in feature extraction and emotion classification understanding for requirements engineering. A demo of the tool is showcased here: https://youtu.be/PFNCbborPuU., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Which design decisions in AI-enabled mobile applications contribute to greener AI?

  • Creus Castanyer, Roger
  • Martínez Fernández, Silverio Juan|||0000-0001-9928-133X
  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
Background: The construction, evolution and usage of complex artificial intelligence (AI) models demand expensive computational resources. While currently available high-performance computing environments support well this complexity, the deployment of AI models in mobile devices, which is an increasing trend, is challenging. Mobile applications consist of environments with low computational resources and hence imply limitations in the design decisions during the AI-enabled software engineering lifecycle that balance the trade-off between the accuracy and the complexity of the mobile applications. Objective: Our objective is to systematically assess the trade-off between accuracy and complexity when deploying complex AI models (e.g. neural networks) to mobile devices, which have an implicit resource limitation. We aim to cover (i) the impact of the design decisions on the achievement of high-accuracy and low resource-consumption implementations; and (ii) the validation of profiling tools for systematically promoting greener AI. In this way, we aim to provide a quantitative analysis of the performance of AI-enabled applications in operation with respect to their design decisions. Method: This confirmatory registered report consists of a plan to conduct an empirical study to quantify the implications of the design decisions on AI-enabled applications performance and to report experiences of the end-to-end AI-enabled software engineering lifecycle. Concretely, we will implement both image-based and language-based neural networks in mobile applications to solve multiple image classification and text classification problems on different benchmark datasets. For that, we make use of the Android Studio and Unity3D frameworks for building mobile applications that work over Android; and TensorFlow Lite and Open Neural Network Exchange (ONNX) file formats for deploying the AI models in the applications. Overall, we plan to model the accuracy and complexity of AI-enabled applications in operation with respect to their design decisions and will provide tools for allowing practitioners to gain consciousness of the quantitative relationship between the design decisions and the green characteristics of study. Additionally, we will provide experiences in the end-to-end AI-enabled software lifecycle and discuss the challenges found, tools and practices for practitioners. Finally, we will provide an open-source data repository following the ESEM open science practices and containing all the experimentation, analysis and reports in our study., This work has been partially supported by the DOGO4ML Spanish research project (ref. PID2020-117191RB-I00) and by the the "Beatriz Galindo" Spanish Program BEAGAL18/00064., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

T-FREX: a transformer-based feature extraction method from mobile app reviews

  • Motger de la Encarnación, Joaquim|||0000-0002-4896-7515
  • Miaschi, Alessio
  • Dell’Orletta, Felice
  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
  • Marco Gómez, Jordi|||0000-0002-0078-7929
Mobile app reviews are a large-scale data source for software-related knowledge generation activities, including software maintenance, evolution and feedback analysis. Effective extraction of features (i.e., functionalities or characteristics) from these reviews is key to support analysis on the acceptance of these features, identification of relevant new feature requests and prioritization of feature development, among others. Traditional methods focus on syntactic pattern-based approaches, typically context-agnostic, evaluated on a closed set of apps, difficult to replicate and limited to a reduced set and domain of apps. Mean-while, the pervasiveness of Large Language Models (LLMs) based on the Transformer architecture in software engineering tasks lays the groundwork for empirical evaluation of the performance of these models to support feature extraction. In this study, we present T-FREX, a Transformer-based, fully automatic approach for mobile app review feature extraction. First, we collect a set of ground truth features from users in a real crowdsourced software recommendation platform and transfer them automatically into a dataset of app reviews. Then, we use this newly created dataset to fine-tune multiple LLMs on a named entity recognition task under different data configurations. We assess the performance of T- FREX with respect to this ground truth, and we complement our analysis by comparing T- FREX with a baseline method from the field. Finally, we assess the quality of new features predicted by T- FREX through an external human evaluation. Results show that T- FREX outperforms on average the traditional syntactic-based method, especially when discovering new features from a domain for which the model has been fine-tuned., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovacion under project I funding scheme PID2020-117191RB-100 I AEI/10.13039/501100011033. Alessio Miaschi and Felice Dell’ Orletta have also been supported by the PNRR project FAIR - Future AI Research (PE00000013), under the NRRP MUR program funded by the NextGenerationEU., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

DOGO4ML: Development, operation and data governance for ML-based software systems

  • Ayala Martínez, Claudia Patricia|||0000-0002-6262-3698
  • Bilalli, Besim|||0000-0002-0575-2389
  • Gómez Seoane, Cristina|||0000-0002-3872-0439
  • Martínez Fernández, Silverio Juan|||0000-0001-9928-133X
Machine Learning based Software Systems (MLSS) are becoming increasingly pervasive in today’s society and can be found in virtually every domain. Building MLSS is challenging due to their interdisciplinary nature. MLSS engineering encompasses multiple disciplines, of which Data Engineering and Software Engineering appear as most relevant. The DOGO4ML project aims at reconciling these two disciplines for providing a holistic end-to-end framework to develop, operate and govern MLSS and their data. It proposes to combine and intertwine two software cycles: the DataOps and the DevOps lifecycles. The DataOps lifecycle manages the complexity of dealing with the big data needed by ML models, while the DevOps lifecycle is in charge of building the system that embeds these models. In this paper, we present the main vision and goals of the project as well as its expected contributions and outcomes. Although the project is in its initial stage, the progress of the research undertaken so far is detailed., This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Software analytics tools: an intentional view

  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
Software analytic tools consume big amounts of data coming from either (or both) the software development process or the system usage and aggregate them into indicators which are rendered to different types of stakeholders, also offering them a portfolio of techniques and capabilities such as what-if analysis, prediction and alerts. Precisely, the variety of stakeholders and the different goals they pursue justifies the convenience of performing an intentional analysis of the use of software analytics tools. With this aim, we first enumerate the different stakeholders and identify their intentional relationships with software analytics tools in the form of dependencies. Then, we focus on one particular stakeholder, namely the requirements engineer, and identify further intentional elements represented in a strategic rationale model. The resulting model provides an abstract view of the domain which may help stakeholders when deciding on the adoption of software analytic tools in their particular context., This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

MApp-KG: Mobile app knowledge graph for document-based feature knowledge generation

  • Motger de la Encarnación, Joaquim|||0000-0002-4896-7515
  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
  • Marco Gómez, Jordi|||0000-0002-0078-7929
Mobile app repositories serve as large-scale crowdsourced information systems used for various document-based software engineering tasks, leveraging product descriptions, user reviews, and other natural language documents. Particularly, feature extraction (i.e., identifying functionalities or capabilities of a mobile app mentioned in these documents) is key for product recommendation, topic modelling, and feedback analysis. However, researchers often face domain-specific challenges in mining these repositories, including the integration of heterogeneous data sources, large-scale data collection, normalization and ground-truth generation for feature-oriented tasks. In this paper, we introduce MApp-KG, a combination of software resources and data artefacts in the field of mobile app repositories aimed at supporting feature-oriented knowledge generation tasks. Our contribution provides a framework for automatically constructing a knowledge graph that models a domain-specific catalog of natural language documents related to mobile applications. We distribute MApp-KG through a public triplestore, enabling its immediate use for future research and replication of our findings., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Hyppo: efficient discovery and execution of data science pipelines in collaborative environments

  • Kontaxakis, Antonios
  • Sacharidis, Dimitris
  • Simitsis, Alkis
  • Abelló Gamazo, Alberto|||0000-0002-3223-2186
  • Nadal Francesch, Sergi|||0000-0002-8565-952X
In exploratory data science and machine learning (ML), developing an effective and efficient solution involves the exploration of numerous pipelines per dataset, considering various combinations of data preprocessing, feature selection, model selection, evaluation metrics, etc. This is an iterative process of trial and error, where a pipeline is revised and refined until a satisfactory level of performance is achieved. Such a process can be substantially improved by leveraging collaboration opportunities and automated pipeline optimization. We demonstrate Hyppo, a scalable collaboration system designed for interactive discovery and automated optimization, resulting in a system that supports evaluating and sharing pipelines with minimal overhead. Hyppo takes advantage of numerous optimization possibilities as exploiting equivalences among tasks, sharing computations, artifact materialization, and reuse from past executions to derive better execution plans. We showcase Hyppo with open data and pipelines from popular public code repositories and catalogs. Demo video: https://youtu.be/iilPV_xtXwM, This work has been partially supported by the H2020-MSCA-ITN-2020 DEDS (GA.955895), the European Union’s Horizon Europe Research and Innovation programmes DataGEMS (GA.101188 416), CREXDATA (GA.101092749), ExtremeXP (GA.101093164), and the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191 RB-I00/AEI/10.13039/50 1100011033 (DOGO4ML)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Chatbots4Mobile: Feature-oriented knowledge base generation using natural language

  • Motger de la Encarnación, Joaquim|||0000-0002-4896-7515
  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
  • Marco Gómez, Jordi|||0000-0002-0078-7929
Chatbots4Mobile is a research project from the GESSI research group (UPC-BarcelonaTech) which aims at designing and developing a task oriented, knowledge based conversational agent to support mobile users in the process of managing and integrating the functionalities exposed by their own application portfolio. To support the design of the required knowledge base, the project focuses on the application of Natural Language Processing (NLP) techniques to infer extended knowledge about the features exposed by a subset of mobile applications, including feature extraction from app-related documents, syntactic and semantic similarity analysis between features, and intent/entity classification focused on functionality identification from user requests. As next steps, we are focusing on the evaluation of embedded linguistic knowledge in large language models, as well as the application of granular sentiment analysis techniques to discern biased documents and process sentiment-based user feedback., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

How and why false denial constraints are discovered

  • Martín Garcia, Albert|||0009-0003-9967-9331
  • Cunha de Almeida, Eduardo
  • Romero Moral, Óscar|||0000-0001-6350-8328
  • Queralt Calafat, Anna|||0000-0003-2782-2955
Denial Constraints (DCs) are a flexible formalism to express many types of data rules, making them a widely adopted tool for many applications. This flexibility led to the development of numerous algorithms to automatically discover DCs directly from data. However, few studies have been conducted on the quality of the discovered DCs. We experimentally quantify the lack of quality in the results obtained by state-of-the-art algorithms, showing how the proportion of discovered DCs that are false is rarely below 95%. We hypothesize that the common source of these erroneous DCs stems from the adoption of the current DC validity definition. We use a statistical approach to explain the mechanism leading to these results, and propose a redefinition of DC validity properties to avoid the acceptance of false DCs. We validate this redefinition experimentally, showing that it exclusively accepts true constraints of the data, and is reliable enough to discover DCs missed by domain experts. Additionally, we provide curated sets of golden DCs for each dataset used in our study, those generated by domain experts and those discovered using our approach., This work is supported by the Horizon Europe Programme under GA.101135513 (CyclOps) and the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RB-I00 / AEI/10.13039/ 501100011033 (DOGO4ML). Anna Queralt is a Serra-Húnter fellow. E. Almeida is funded by the CNPQ grants 302909/2022-2 and 444192/2024-7. Albert Martin is funded by the predoctoral program AGAUR-FI grants (2025 FI-1 00967) Joan Oró, which is backed by the Secretariat of Universities and Research of the Department of Research and Universities of the Generalitat of Catalonia, as well as the European Social Plus Fund., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Using metrics for code smells of ML pipelines

  • Costal Costa, Dolors|||0000-0002-7340-0414
  • Gómez Seoane, Cristina|||0000-0002-3872-0439
  • Rey Juárez, Santiago del|||0000-0003-4979-414X
  • Martínez Fernández, Silverio Juan|||0000-0001-9928-133X
© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. https://dx.doi.org/0.1109/ICSA-C63560.2024.00055, ML pipelines, as key components of ML systems, shall be developed following quality assurance techniques. Unfortunately, it is often the case in which they present maintainability issues, due to the experimental nature of data collection and ML model construction. In this work, we perform a first evaluation of a set of metrics, proposed in previous research, for measuring the presence of code smells related to maintainability, in ML pipeline application examples. Moreover, we provide the lessons learnt and insights gained from this evaluation., This work has been funded by the National Spanish Program for Research Aimed at the Challenges of Society 2020. The project and funding identifiers are PID2020-117191RB-I00/ AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Hyppo: using equivalences to optimize pipelines in exploratory machine learning

  • Kontaxakis, Antonios
  • Sacharidis, Dimitris
  • Simitsis, Alkis
  • Abelló Gamazo, Alberto|||0000-0002-3223-2186
  • Nadal Francesch, Sergi|||0000-0002-8565-952X
We present HYPPO, a novel system to optimize pipelines encountered in exploratory machine learning. HYPPO exploits alternative computational paths of artifacts from past executions to derive better execution plans while reusing materialized artifacts. Adding alternative computations introduces new challenges for exploratory machine learning regarding workload representation, system architecture, and optimal execution plan generation. To this end, we present a novel workload representation based on directed hypergraphs, and we formulate the problem of discovering the optimal execution plan as a search problem over directed hypergraphs and that of selecting artifacts to materialize as an optimization problem. A thorough experimental evaluation shows that HYPPO results in plans that are typically one order (up to two orders) of magnitude faster and cheaper than the non-optimized pipeline and considerably (up to one order of magnitude) faster and cheaper than plans generated by the state of the art when materializing artifacts is possible. Lastly, our evaluation reveals that HYPPO reduces the cost by 3–4× even when materialization cannot be exploited., This work has been partially supported by the H2020-MSCAITN-2020 DEDS (GA.955895), the EU-HORIZON programmes FAIR-CORE4EOSC (GA.101057264), CREXDATA (GA.101092749), ExtremeXP (GA.101093164), and the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RB-I00/AEI/10.13039/501100011033 (DOGO4ML)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Energy efficiency of training neural network architectures: an empirical study

  • Xu, Yinlena
  • Martínez Fernández, Silverio Juan|||0000-0001-9928-133X
  • Martínez Martínez, Matías-Sebastián|||0000-0002-2945-866X
  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
The evaluation of Deep Learning (DL) models has traditionally focused on criteria such as accuracy, F1 score, and related measures. The increasing availability of high computational power environments allows the creation of deeper and more complex models. However, the computations needed to train such models entail a large carbon footprint. In this work, we study the relations between DL model architectures and their environmental impact in terms of energy consumed and CO2 emissions produced during training by means of an empirical study using Deep Convolutional Neural Networks. Concretely, we study: (i) the impact of the architecture and the location where the computations are hosted on the energy consumption and emissions produced; (ii) the trade-off between accuracy and energy efficiency; and (iii) the difference on the method of measurement of the energy consumed using software-based and hardware-based tools., This work has been supported by the Spanish project PID2020-117191RB-I00 funded by MCIN/AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Goal-oriented models for teaching and understanding data structures

  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
  • Ruiz, Marcela
Most computer science curricula include a compulsory course on data structures. Students are prone to memorise facts about data structures instead of understanding the essence of underlying concepts. This can be explained by the fact that learning the basics of each data structure, the difference with each other, and the adequacy of each of them to the most appropriate context of use, is far from trivial. This paper explores the idea of providing adequate levels of abstractions to describe data structures from an intentional point of view. Our hypothesis is that adopting a goal-oriented perspective could emphasise the main goals of each data structure, its qualities, and its relationships with the potential context of use. Following this hypothesis, in this paper we present the use of iStar2.0 to teach and understand data structures. We conducted a comparative quasi-experiment with undergraduate students to evaluate the effectiveness of the approach. Significant results show the great potential of goal modeling for teaching technical courses like data structures. We conclude this paper by reflecting on further teaching and conceptual modeling research to be conducted in this field., This work has been partially supported by the by the DOGO4ML Spanish research project (ref. PID2020-117191RB-I00), the Digitalization Initiative of the Canton of Zü-rich (DIZH), and ZHAW Digital., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Discovery of semantic non-syntactic joins

  • Maynou Yelamos, Marc|||0009-0004-3908-4392
  • Nadal Francesch, Sergi|||0000-0002-8565-952X
Data discovery is an essential step in the data integration pipeline involving finding datasets whose combined information provides relevant insights. Discovering joinable attributes requires assessing the closeness of the semantic concepts that two attributes represent, which is highly sensitive and dependent on the chosen similarity metric. The state of the art commonly approaches this task from a syntactic perspective, this is, performing comparisons based on the data values or on direct transformations (e.g., via hash functions). These approaches suffice when the two sets of instances share the same syntactic representation, but fail to detect cases in which the same semantic concept is represented by different sets of values, which we refer as semantic non-syntactic joins. This is a relevant problem in data lake scenarios, when the underlying datasets present high heterogeneity and lack of standardization. To that end, in this paper, we propose an empirical approach to detect semantic non-syntactic joins, which leverages, simultaneously, syntactic and semantic measurements of the data. We demonstrate that our approach is effective in detecting such kind of joins., Marc Maynou is supported by the EU’s Horizon Programme call, under Grant Agreements No. 101093164 (ExtremeXP), and Sergi Nadal is partially supported by the DOGO4ML project, funded by the Spanish Ministerio de Ciencia e Innovación under the funding scheme PID2020-117191RB-I00 / AEI / 10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Measuring discord among multidimensional data sources

  • Abelló Gamazo, Alberto|||0000-0002-3223-2186
  • Cheney, James
Data integration is a classical problem in databases, typically decomposed into schema matching, entity matching and record merging. To solve the latter, it is mostly assumed that ground truth can be determined, either as master data or from user feedback. However, in many cases, this is not the case because firstly the merging processes cannot be accurate enough, and also the data gathering processes in the different sources are simply imperfect and cannot provide high quality data. Instead of enforcing consistency, we propose to evaluate how concordant or discordant sources are as a measure of trustworthiness (the more discordant are the sources, the less we can trust their data). Thus, we define the discord measurement problem in which given a set of uncertain raw observations or aggregate results (such as case/hospitalization/death data relevant to COVID-19) and information on the alignment of different data (for example, cases and deaths), we wish to assess whether the different sources are concordant, or if not, measure how discordant they are., The work of Alberto Abelló has been done under project PID2020- 117191RB-I00 funded by MCIN/ AEI /10.13039/501100011033. The work of James Cheney was supported by ERC Consolidator Grant Skye (grant number 682315)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Mining app reviews for user feedback analysis in requirements engineering: a project report

  • Motger de la Encarnación, Joaquim|||0000-0002-4896-7515
  • Oriol Hilari, Marc|||0000-0003-1928-7024
  • Tiessler Aguirre, Max
  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
  • Marco Gómez, Jordi|||0000-0002-0078-7929
Mining app reviews has emerged as a valuable practice in requirements engineering, providing insights into feature usage trends, user satisfaction, and emerging software issues. While recent advances in natural language processing have enhanced review analysis, challenges persist in feature extraction, sentiment ambiguity, and the scalability of automated methods, among others. This project report presents our research efforts in app review mining, focusing on methodological, software-based, and data-driven contributions. We explore both supervised and unsupervised learning approaches, leveraging large language models for key tasks such as feature identification, competition analysis, and emotion extraction. Additionally, we develop open-source tools and datasets to support reproducibility and adoption of our methods. Our findings highlight the potential of large language models in automating user feedback analysis while identifying gaps that require further research, particularly in addressing model reliability and evaluation challenges., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Inclusion and exclusion criteria in software engineering tertiary studies: A systematic mapping and emerging framework

  • Costal Costa, Dolors|||0000-0002-7340-0414
  • Farré Tost, Carles|||0000-0001-5814-3782
  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
  • Quer, Carme|||0000-0002-9000-6371
Context: Tertiary studies in software engineering (TS@SE) are widely used to synthesise evidence on a research topic systematically. As part of their protocol, TS@SE define inclusion and exclusion criteria (IC/EC) aimed at selecting those secondary studies (SS) to be included in the analysis. Aims: To provide a state of the art on the definition and application of IC/EC in TS@SE, and from the results of this analysis, we outline an emerging framework, TSICEC, to be used by SE researchers. Method: To provide the state of the art, we conducted a systematic mapping (SM) combining automatic search and snowballing over the body of SE scientific literature, which led to 50 papers after application of our own IC/EC. The extracted data was synthesised using content analysis. The results were used to define a first version of TSICEC. Results: The SM resulted in a coding schema, and a thorough analysis of the selected papers on the basis of this coding. Our TSICEC framework includes guidelines for the definition of IC/EC in TS@SE. Conclusion: This paper is a step forward establishing a foundation for researchers in two ways. As authors, understanding the different possibilities to define IC/EC and apply them to select SS. As readers, having an instrument to understand the methodological rigor upon which TS@SE may claim their findings., This work has been partially supported by the DOGO4ML Spanish research project (ref. PID2020-117191RB-I00)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Insights on the use of software design principles in machine learning pipelines

  • López Cuesta, Lidia|||0000-0002-6901-9223
  • Gómez Seoane, Cristina|||0000-0002-3872-0439
  • Ayala Martínez, Claudia Patricia|||0000-0002-6262-3698
The rapid expansion of Artificial Intelligence has driven a surge in the development of Machine Learning (ML) pipelines, essential for constructing and maintaining ML models. Despite the growing recognition of the importance of high-quality ML pipelines, there is limited actionable guidance on improving pipelines quality at the implementation level. The Object-Oriented (OO) paradigm and software design principles have been used in software engineering to enhance software system quality. In this context, this work explores the coverage of existing ML pipelines in relation to OO concepts and software design principles. Our findings reveal a limited use of OO mechanisms and minimal adherence to software design principles. Specifically, issues such as poor code organization and deficient levels of coupling and cohesion were observed. Our results can shed some light on the development of strategies to overcome these eficiencies for improving ML pipelines quality., This work was supported by the “Spanish Ministerio de Ciencia e Innovación”, under project PID2020-117191RB-I00, funding scheme AEI/https://doi.org/10.13039/ 501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Adaptive task-oriented chatbots using feature-based knowledge bases

  • Campàs Gené, Carla
  • Motger de la Encarnación, Joaquim|||0000-0002-4896-7515
  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
  • Marco Gómez, Jordi|||0000-0002-0078-7929
Task-oriented chatbots relying on a knowledge base for domain-specific content exploitation have been largely addressed in research and industry applications. Despite this, multiple challenges remain to be fully conquered, including adaptive knowledge mechanisms, personalization for user-specific demands, and composite intent resolution. To address these challenges, in this paper, we present a work-in-progress summary of a task-oriented, knowledge-based chatbot in the field of mobile software ecosystems. The chatbot is designed to assist users in the combined use of multiple features from different applications. The proposed knowledge base and the machine learning pipeline supporting the chatbot technical core are designed to: (i) effectively use user context, (ii) process runtime feedback, (iii) use user historical data, and (iv) automatically infer slot values and dependent actions. With this report, we expect to lay the groundwork for future development stages and user validation studies., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Data-driven agile requirements elicitation through the lenses of situational method engineering

  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
  • Henriksson, Aron
  • Ralyté, Jolita
  • Zdravkovic, Jelena
Ubiquitous digitalization has led to the continuous generation of large amounts of digital data, both in organizations and in society at large. In the requirements engineering community, there has been a growing interest in considering digital data as new sources for requirements elicitation, in addition to stake-holders. The volume, dynamics, and variety of data makes iterative requirements elicitation increasingly continuous, but also unstructured and complex, which current agile methods are unable to consider and manage in a systematic and efficient manner. There is also the need to support software evolution by enabling a synergy of stakeholder-driven requirements elicitation and management with data-driven approaches. In this study, we propose extension of agile requirements elicitation by applying situational method engineering. The research is grounded on two studies in the business domains of video games and online banking., The work presented in this paper is partially funded by the DOGO4ML Spanish research project, PID2020-117191RB-I00., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Towards continuous experiment-driven MLOps

  • Rajenthiram, Keerthiga
  • Abdullah, Milad
  • Gerostathopoulos, Ilias
  • Hnetynka, Petr
  • Bureš, Tomáš
  • Pons Recasens, Gerard|||0000-0003-2225-3255
  • Bilalli, Besim|||0000-0002-0575-2389
  • Queralt Calafat, Anna|||0000-0003-2782-2955
Despite advancements in MLOps and AutoML, ML development still remains challenging for data scientists. First, there is poor support for and limited control over optimizing and evolving ML models. Second, there is lack of efficient mechanisms for continuous evolution of ML models which would leverage the knowledge gained in previous optimizations of the same or different models. We propose an experiment-driven MLOps approach which tackles these problems. Our approach relies on the concept of an experiment, which embodies a fully controllable optimization process. It introduces full traceability and repeatability to the optimization process, allows humans to be in full control of it, and enables continuous improvement of the ML system. Importantly, it also establishes knowledge, which is carried over and built across a series of experiments and allows for improving the efficiency of experimentation over time. We demonstrate our approach through its realization and application in the ExtremeXp11https://extremexp.eu/ project (Horizon Europe)., This work has been partially supported by the EU project ExtremeXP grant agreement 101093164, by Charles University institutional funding SVV 260698, and by the Spanish Ministerio de Ciencia e Innovación under project PID2020- 117191RB-I00/ AEI/10.13039/501100011033 (DOGO4ML)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Mitigating data sparsity in integrated data through text conceptualization

  • Rahman, Md Ataur|||0000-0002-9092-8771
  • Nadal Francesch, Sergi|||0000-0002-8565-952X
  • Romero Moral, Óscar|||0000-0001-6350-8328
  • Sacharidis, Dimitris
We study the data sparsity problem for data generated from an integration system. We approach the problem from a textual information extraction perspective and propose to conceptualize external documents using the concepts in the integrated schema. We present THOR, a novel system that, unlike related approaches, neither relies on complex rules nor models trained with large annotated corpus, but on the integrated data and its schema without the need for human annotations. An extensive evaluation on the text conceptualization task demonstrates the superiority of our approach in terms of F1-score, effort and use of resources over the state-of-the-art language models., Md Ataur Rahman is funded by the EC Horizon 2020 research and innovation programme (DEDS: grant agreement No 955895). Sergi Nadal and Oscar Romero are funded by the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RB-I00/AEI/10.13039/501100011033 (DOGO4ML) and the EC Horizon Europe programme (ExtremeXP: grant agreement No 101093164)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

A data-science pipeline to enable the interpretability of many-objective feature selection

  • Njoku, Uchechukwu Fortune|||0000-0002-2599-9645
  • Abelló Gamazo, Alberto|||0000-0002-3223-2186
  • Bilalli, Besim|||0000-0002-0575-2389
  • Bontempi, Gianluca
Many-Objective Feature Selection (MOFS) approaches use four or more objectives to determine the relevance of a subset of features in a supervised learning task. As a consequence, MOFS typically returns a large set of non-dominated solutions, which have to be assessed by the data scientist in order to proceed with the final choice. Given the multi-variate nature of the assessment, which may include objectives (e.g., fairness) unrelated to predictive accuracy, this step is often not straightforward and suffers from the lack of existing tools. For instance, it is common to make use of a tabular presentation of the solutions, which provides little information about the trade-offs and the relationships between objectives over the set of solutions. Adopting a GA-based MOFS with six objectives (number of selected features, balanced accuracy, F1-Score, variance inflation factor, statistical parity, and equalised odds) for two feature selection tasks, this paper illustrates the complex challenge of assessing MOFS results and the need for a methodology to aid and justify the final choice of a solution., The project leading to this publication has received funding from the European Union’s Horizon 2020 research and innovation programme (G.A. No. 955895). A. Abelló and B. Bilalli are funded by the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RBI00, funding scheme AEI/10.13039/501100011033. G. Bontempi is supported by the Service Public de Wallonie Recherche under G.A. No. 2010235–ARIAC by DigitalWallonia4.ai. Computational resources have been provided by the Consortium des Équipements de Calcul Intensif, funded by the Fonds de la Recherche Scientifique de Belgique (G.A. No. 2.5020.11) and Walloon Region., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Mobile feature-oriented knowledge base generation using knowledge graphs

  • Motger de la Encarnación, Joaquim|||0000-0002-4896-7515
  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
  • Marco Gómez, Jordi|||0000-0002-0078-7929
Knowledge bases are centralized repositories used for developing knowledge-oriented information systems. They are essential for adaptive, specialized knowledge in dialogue systems, supporting up-to-date domain-specific discussions with users. However, designing large-scale knowledge bases presents multiple challenges in data collection and knowledge exploitation. Knowledge graphs provide various research opportunities by integrating decentralized data and generating advanced knowledge. Our contribution presents a knowledge base in the form of a knowledge graph for extended knowledge generation for mobile apps and features extracted from related natural language documents. Our work encompasses the knowledge graph completion and deductive and inductive knowledge requests. We evaluated the effectiveness and performance of these knowledge strategies, which can be used as on-demand knowledge requests used by third-party software systems., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Leveraging requirements elicitation through software requirement patterns and LLMs

  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
  • Gnesi, Stefania
  • Paccosi, Federico
  • Quer, Carme|||0000-0002-9000-6371
  • Semini, Laura
Software requirement patterns (SRPs) is one of the many techniques that contribute to requirements elicitation. At this respect, the emergence of large language models (LLMs) opens the door to cost-effective strategies to create and use SRPs. Still, the stochastic nature of LLMs threatens the inherent quality of requirements reuse and consequently, that of the elicitation process. [Question/problem] In this scientific evaluation paper, we investigate whether and how LLMs can be used in order to create an SRP catalogue and elicit requirements from it. [Principal ideas/results] SRPs can be effectively extracted by querying an LLM through appropriate prompts, but still expert assessment is key in order to deliver the best results. LLM-driven generation of questions to stakeholders for eliciting requirements from these SRPs is feasible but suffers from deficiencies such as excessive number of repetitions and out of scope requirements. [Contribution] We show that (1) LLMs can be embedded into the requirements elicitation process through a pattern instantiation-based strategy, but at the same time (2) the current state of LLM technologies requires expert assessment at a large extent., This paper has been partially funded by the Spanish Ministerio de Ciencia e Innovación under project/funding scheme PID2020-117191RBI00/AEI/10.13039/501100011033, by the Italian MIUR, under project PRIN 2022 STENDHAL and PNRR Project Securing sOftware Platforms.
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

A data quality framework for graph-based virtual data integration systems

  • Li, Yalei
  • Nadal Francesch, Sergi|||0000-0002-8565-952X
  • Romero Moral, Óscar|||0000-0001-6350-8328
Data Quality (DQ) plays a critical role in data integration. Up to now, DQ has mostly been addressed from a single database perspective. Popular DQ frameworks rely on Integrity Constraints (IC) to enforce valid application semantics, which lead to the Denial Constraint (DC) formalism which models a broad range of ICs in real-world applications. Yet, current approaches are rather monolithic, considering a single database and do not suit data integration scenarios. In this paper, we address DQ for data integration systems. Specifically, we extend virtual data integration systems to elicit DCs from disparate data sources to be integrated, using DC-related state-of-the-art, and propagate them to the integrated schema (global DCs). Then, we propose a method to manage global DCs and identify (i) minimal DCs and (ii) potential clashes between them., This work was partly supported by the DOGO4ML project, funded by the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RB-I00. Sergi Nadal is partly supported by the Spanish Ministerio de Ciencia e Innovación, as well as the European Union - NextGenerationEU, under project FJC2020-045809-I., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

What about emotions? Guiding fine-grained emotion extraction from mobile app reviews

  • Motger de la Encarnación, Joaquim|||0000-0002-4896-7515
  • Oriol Hilari, Marc|||0000-0003-1928-7024
  • Tiessler Aguirre, Max
  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
  • Marco Gómez, Jordi|||0000-0002-0078-7929
Opinion mining plays a vital role in analysing user feedback and extracting insights from textual data. While most research focuses on sentiment polarity (e.g., positive, negative, neutral), fine-grained emotion classification in app reviews remains underexplored. Fine-grained emotion classification is thus needed to better understand users’ affective responses and support downstream tasks such as feature-emotion analysis, user-oriented release planning, and issue triaging. This paper addresses this gap by identifying and addressing the challenges and limitations in fine-grained emotion analysis in the context of app reviews. Our study adapts Plutchik’s emotion taxonomy to app reviews by developing a structured annotation framework and dataset. Through an iterative human annotation process, we define clear annotation guidelines and document key challenges in emotion classification. Additionally, we evaluate the feasibility of automating emotion annotation using large language models, assessing their cost-effectiveness and agreement with human-labelled data. Our findings reveal that while large language models significantly reduce manual effort and maintain substantial agreement with human annotators, full automation remains challenging due to the complexity of emotional interpretation. This work contributes to opinion mining in requirements engineering by providing structured guidelines, an annotated dataset, and insights for developing automated pipelines to capture the complexity of emotions in app reviews., This work has been supported by funding from the HIVEMIND project – Horizon Europe call HORIZON-CL4-2024- DIGITAL-EMERGING-01 under Grant Agreement Number 101189745. This paper has been funded by the Spanish Ministerio de Ciencia e Innovacióin under project/funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Architectural decisions in AI-based systems: an ontological view

  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
  • Martínez Fernández, Silverio Juan|||0000-0001-9928-133X
  • Ayala Martínez, Claudia Patricia|||0000-0002-6262-3698
  • Gómez Seoane, Cristina|||0000-0002-3872-0439
Architecting AI-based systems entails making some decisions that are particular to this type of systems. Therefore, it becomes necessary to gather all necessary knowledge to inform such decisions, and to articulate this knowledge in a form that facilitates knowledge transfer among different AI projects. In this exploratory paper, we first present the main results of a literature survey in the area, and then we propose a preliminary ontology for architectural decision making, which we exemplify using a subset of the papers selected in the literature review. In the discussion, we remark on the variety of decision types and system contexts, highlighting the need to further investigate the current state of research and practice in this area. Besides, we summarize our plans to move along this research area by widening the literature review and incorporating more AI-related concepts to this first version of the ontology., This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Web API change-proneness prediction

  • Koçi, Rediana|||0000-0003-3367-950X
  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
  • Jovanovic, Petar|||0000-0003-4635-6646
  • Abelló Gamazo, Alberto|||0000-0002-3223-2186
Change-proneness of software artifacts has been mainly related to the design characteristics and their previous history of changes. While these two aspects are essential and contribute significantly to the prediction, they leave out a critical factor: how the artifacts are being used. In the context of web APIs, consumers represent one of the main drivers of the change. Therefore, we propose a methodology for predicting the change-proneness of web API endpoint interfaces, taking into account not only design and change history but also their usage. Since the evolution of web APIs is and should be usage-driven, the way consumers use an API affects the future changes implemented by providers. Consequently, consumers' usage behavior contains essential information that contributes to identifying endpoints that are more prone to change. By considering the reasons behind changes, we introduce a set of metrics comprising design and usage aspects to be used as variables in prediction. To demonstrate the usefulness of the approach we perform an initial evaluation using a real-world web API. We quantify the introduced metrics using web API documentation, code, and usage logs in order to build a classifier able to predict with 82 % accuracy if an endpoint will change based on its design, history of changes, and usage characteristics., This paper has been funded by the Spanish Ministerio de Ciencia e Innovacion under project/funding scheme PID2020-117191RB-I00/AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Merging datasets for emotion analysis

  • de Arriba Serra, Ariadna
  • Oriol Hilari, Marc|||0000-0003-1928-7024
  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
Context. Applying sentiment analysis is in general a laborious task. Furthermore, if we add the task of getting a good quality dataset with balanced distribution and enough samples, the job becomes more complicated. Objective. We want to find out whether merging compatible datasets improves emotion analysis based on machine learning (ML) techniques, compared to the original, individual datasets. Method. We obtained two datasets with Covid-19-related tweets written in Spanish, and then built from them two new datasets combining the original ones with different consolidation of balance. We analyzed the results according to precision, recall, F1-score and accuracy. Results. The results obtained show that merging two datasets can improve the performance of ML models, particularly the F1-score, when the merging process follows a strategy that optimizes the balance of the resulting dataset. Conclusions. Merging two datasets can improve the performance of ML models for emotion analysis, whilst saving resources for labeling training data. This might be especially useful for several software engineering activities that leverage on ML-based emotion analysis techniques., This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Metrics for code smells of ML pipelines

  • Costal Costa, Dolors|||0000-0002-7340-0414
  • Gómez Seoane, Cristina|||0000-0002-3872-0439
  • Martínez Fernández, Silverio Juan|||0000-0001-9928-133X
ML pipelines, as key components of ML systems, shall be developed following quality assurance techniques. Unfortunately, it is often the case in which they present maintainability issues, due to the experimentatal nature of data collection and ML model construction. To address this problem, this work in progress proposes initial metrics to measure the presence of code smells in ML pipelines. These metrics reflect good software engineering practices for code quality of ML pipelines., This work was supported by the “Spanish Ministerio de Ciencia e Innovación” under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

There is no data science without data governance: a proposal based on knowledge graphs

  • Bilalli, Besim|||0000-0002-0575-2389
  • Jovanovic, Petar|||0000-0003-4635-6646
  • Nadal Francesch, Sergi|||0000-0002-8565-952X
  • Queralt Calafat, Anna|||0000-0003-2782-2955
  • Romero Moral, Óscar|||0000-0001-6350-8328
Data Science and data-driven Artificial Intelligence are here to stay and they are expected to further transform the current global economy. From a technical point of view, there is an overall agreement that disciplines based on data require to combine data engineering and data analysis skills, but the fact is that data engineering is nowadays trailing and catching up with the rapid changes in the data analysis landscape. To unleash the real power of data, data-centric systems must be professionalized, i.e., operationalized and systematized, so that repetitive, time-consuming and error-prone tasks are automated. To such end, we propose our vision on next generation data governance for data-centric systems based on knowledge graphs. We claim that without the knowledge embedded in the data governance layer, Data Science will not unleash its potential., This work is supported by the Horizon Europe Programme under GA.101135513 (CYCLOPS) and GA. 101093164 (ExtremeXP) and the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RBI00 / AEI/10.13039/501100011033 (DOGO4ML)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Knowledge graphs for enhancing large language models in entity disambiguation

  • Pons Recasens, Gerard|||0000-0003-2225-3255
  • Bilalli, Besim|||0000-0002-0575-2389
  • Queralt Calafat, Anna|||0000-0003-2782-2955
Recent advances in Large Language Models (LLMs) have positioned them as a prominent solution for Natural Language Processing tasks. Notably, they can approach these problems in a zero or few-shot manner, thereby eliminating the need for training or fine-tuning task-specific models. However, LLMs face some challenges, including hallucination and the presence of outdated knowledge or missing information from specific domains in the training data. These problems cannot be easily solved by retraining the models with new data as it is a time-consuming and expensive process. To mitigate these issues, Knowledge Graphs (KGs) have been proposed as a structured external source of information to enrich LLMs. With this idea, in this work we use KGs to enhance LLMs for zero-shot Entity Disambiguation (ED). For that purpose, we leverage the hierarchical representation of the entities’ classes in a KG to gradually prune the candidate space as well as the entities’ descriptions to enrich the input prompt with additional factual knowledge. Our evaluation on popular ED datasets shows that the proposed method outperforms non-enhanced and description-only enhanced LLMs, and has a higher degree of adaptability than task-specific models. Furthermore, we conduct an error analysis and discuss the impact of the leveraged KG’s semantic expressivity on the ED performance., This work is supported by the Horizon Europe Programme under GA.101093164 (ExtremeXP) and the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RB-I00/ AEI/10.13039/501100011033 (DOGO4ML). Anna Queralt is a Serra Húnter Fellow., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Applying transfer learning to sentiment analysis in social media

  • de Arriba Serra, Ariadna
  • Oriol Hilari, Marc|||0000-0003-1928-7024
  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
Context: Sentiment analysis is an NLP technique that can be used to automatically obtain the sentiment of a crowd of end-users regarding a software application. However, applying sentiment analysis is a difficult task, especially considering the need of obtaining enough good quality data for training a Machine Learning (ML) model. To address this challenge, transfer learning can help us save time and get better performance results with a limited amount of data. Objective: In this paper, we aim at identifying to which degree transfer learning improves the results of sentiment analysis of messages shared by end-users in social media. Method: We propose a tool-supported framework able to monitor and analyze the sentiment of tweets with different ML models and settings. Using the proposed framework, we apply transfer learning and conduct a set of experiments with multiple datasets. Results: The performance of different ML models with transfer learning from different datasets are obtained and discussed, showing how different factors affect the results, and discussing how they have to be considered when applying transfer learning., This work has been partially supported by the Spanish project DOGO4ML (contract PID2020-117191RB-I00)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Performance analysis of distributed GPU-accelerated task-based workflows

  • Nogueira Lobo de Carvalho, Marcos|||0000-0001-7015-9517
  • Queralt Calafat, Anna|||0000-0003-2782-2955
  • Romero Moral, Óscar|||0000-0001-6350-8328
  • Simitsis, Alkis
  • Tatu, Cristian
  • Badia Sala, Rosa Maria|||0000-0003-2941-5499
We present an empirical approach to identify the key factors affecting the execution performance of task-based workflows on a High Performance Computing (HPC) infrastructure composed of heterogeneous CPU-GPU clusters. Our results reveal that the execution performance in distributed GPU-accelerated task-based workflows highly depends on several interrelated factors regarding the task algorithm, dataset, resources, and system employed. In addition, our analysis identifies key correlations among these factors, presents novel observations, and offers guidelines toward designing an automated method to handle task-based workflows in modern, high-compute capacity, CPU-GPU engines., This work has been partially supported by DEDS (H2020-MSCAITN2020) with grant agreement No. 955895, the EU-HORIZON programme CREXDATA under GA.101092749, the EU-HORIZON programme FAIR-CORE4EOSC under GA.101057264, the EUHORIZON programme EXTREMEXP under GA.101093164, the Spanish Government projects PID2019-107255GB and PID2020117191RB-I00/AEI/10.13039/501100011033andMCIN/AEI/10.13039 /501100011033 (CEX2021-001148-S), and by the Departament de Recerca i Universitats de la Generalitat de Catalunya (2021 SGR 00412, MPiEDist)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

HealthMesh: An architectural framework for federated healthcare data management

  • Bisquert Parés, Aniol
  • Hmimou Ham Man, Achraf|||0009-0004-0448-3406
  • Berral García, Josep Lluís|||0000-0003-3037-3580
  • Gutiérrez Torre, Alberto|||0000-0002-5548-3359
  • Romero Moral, Óscar|||0000-0001-6350-8328
Recently, significant milestones have been achieved in the field of healthcare data analysis. However, alongside these accomplishments, substantial data-related challenges have emerged in the domain of big data management. Modern healthcare projects are no more dealing with a single data repository but many heterogeneous ones and must overcome data variety, privacy and governance issues. Yet, current solutions face a privacy-decentralization trade-off. To address this dual challenge, we introduce HealthMesh, a novel layered architectural framework based on the Data Mesh principles, providing a domaindecentralised paradigm. In addition, the framework incorporates a Semantic Data Model which establishes robust governance, enables interoperability and guarantees policy compliance for all the data assets. To demonstrate the capabilities of the proposed approach, we provide an illustrative example inspired by the use case of the INCISIVE project for breast cancer analytics. Overall, this work makes a significant contribution on collecting key challenges, identifying actors and providing a set of components and guidelines for establishing a holistic framework for the complex field of healthcare data management., The work reported here was supported by the EU-H2020 programme under GA.952179 INCISIVE, Horizon Europe Programme under GA.101095717 (SECURED) and GA.101135513 (CYCLOPS), the Spanish Research State Agency (MICINN/AEI, ERDF/FEDER) under GA.MCIN/AEI/10.13039/501100011033/FEDER (UE DALEST), the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RB-I00 / AEI/10.13039/501100011033 (DOGO4ML), and the Generalitat de Catalunya (AGAUR) 2021-SGR-00478 (CROMAI)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

RE-Miner 2.0: a holistic framework for mining mobile application reviews

  • Tiessler Aguirre, Max
  • Motger de la Encarnación, Joaquim|||0000-0002-4896-7515
In the domain of application stores and marketplaces, user reviews are crucial for supporting multiple requirements engineering tasks. Feature extraction, emotion classification, topic analysis, review type identification, and polarity analysis are key components in requirements prioritization, feedback gathering, and release planning. Empirical evaluation of these techniques is challenging due to data collection complexities and a lack of reproducible methods and available tools. Furthermore, existing studies often focus on isolated tasks, hindering a comprehensive analysis of user perceptions. This paper introduces RE-Miner 2.0, a work-in-progress tool that integrates multiple data extraction and analysis methods in a distributed environment (RE-Miner Ecosystem), enabling a multidimensional and detailed analysis of user feedback. It offers a web-based service for task integration and comparison, supported by persistent storage and a web application that allows analytical visualization of reviews. As a result, RE-Miner 2.0 provides a platform for task integration, replication, and comparison of review mining techniques. Bringing advancements in deep review analysis for requirements engineering. A demo of the tool is showcased here: https://www.youtube.com/watch?v=a11bHSCYqqM., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Impact study of NoSQL refactoring in SkyServer database

  • Gallinucci, Enrico
  • Golfarelli, Matteo
  • Radwan, Wafaa
  • Zarate Calderón, Gabriel
  • Abelló Gamazo, Alberto|||0000-0002-3223-2186
Data modeling in NoSQL databases is notoriously complex and driven by multiple and possibly conflicting requirements. Researchers have proposed methodologies to optimize schema design of a given domain for a given workload; however, due to the agile environment in which NoSQL databases are usually employed, both domain and workload are frequently subject to changes and evolution- possibly neutralizing the benefits of optimization. When this happens, the benefits of a new optimal schema design must be weighed against the costs of migrating the data. In this work, we empirically show the benefits of schema redesign in a real publicly available database. In particular, we identify multiple snapshots (in terms of domain extension and querying workload) in the 20+ years evolution of SkyServer, demonstrate how NoSQL schema optimization at a given time can later backfire, and evaluate the conditions under which data migration becomes beneficial. This takes us to define the foundations and challenges of a framework for continuous NoSQL database refactoring, with the goal of helping DBAs and data engineers decide if, when, and how a NoSQL database should be reconsidered to restore schema design optimality., This work has been partially supported by the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RB-I00/AEI/10.13039/501100011033 (DOGO4ML)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

TransFeatEx: a NLP pipeline for feature extraction

  • Gallego Marfa, Agustí
  • Motger de la Encarnación, Joaquim|||0000-0002-4896-7515
  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
  • Marco Gómez, Jordi|||0000-0002-0078-7929
Mobile app stores provide centralized access to a large data set of mobile app related natural language textual data, including developer’s documentation (e.g., descriptions, changelogs) and user-generated data (e.g., user reviews). Motivated by this context, multiple studies have focused on data-driven elicitation processes for the automatic extraction of the set of features exposed by a catalogue of applications and the inferred, extended knowledge that can be derived from this information. Moreover, with the emerging and generalization of large language models, traditional linguistic-based approaches can be significantly improved by the potential of the knowledge embedded in this kind of models. In this paper, we present TransFeatEx, a NLP-based feature extraction pipeline that combines the use of a RoBERTa-based model with the application of consolidated syntactic and semantic techniques. The pipeline is designed as a customizable, standalone service to be used either as a playground, experimentation tool or as a software component to be embedded into a third-party software system for batch-processing large document corpora. An example of a demo plan is showcased here: https://youtu.be/gfFyi_i_uTw, With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Unveiling competition dynamics in mobile app markets through user reviews

  • Motger de la Encarnación, Joaquim|||0000-0002-4896-7515
  • Franch Gutiérrez, Javier|||0000-0001-9733-8830
  • Gervasi, Vincenzo
  • Marco Gómez, Jordi|||0000-0002-0078-7929
[Context and motivation] User reviews published in mobile app repositories are essential for understanding user satisfaction and engagement within a specific market segment. [Question/problem] Manual analysis of reviews is impractical due to the large data volume, and automated analysis faces challenges like data synthesis and reporting. This complicates the task for app providers in identifying patterns and significant events, especially in assessing the influence of competitor apps. Furthermore, review-based research is mostly limited to a single app or a single app provider, excluding potential competition analysis. Consequently, there is an open research challenge in leveraging user reviews to support cross-app analysis within a specific market segment. [Principal ideas/results] Following a case-study research method in the microblogging app market, we introduce an automatic, novel approach to support mobile app market analysis. Our approach leverages quantitative metrics and event detection techniques based on newly published user reviews. Significant events are proactively identified and summarized by comparing metric deviations with historical baseline indicators within the lifecycle of a mobile app. [Contribution] Results from our case study show empirical evidence of the detection of relevant events within the selected market segment, including software- or release-based events, contextual events and the emergence of new competitors., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project/funding scheme PID2020-117191RBI00/AEI/10.13039/501100011033., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Inclusion and exclusion criteria in software engineering tertiary studies: A systematic mapping and emerging framework

  • Costa, Dolors
  • Farré Tost, Carles
  • Franch Gutiérrez, Javier
  • Quer, Carme
Context: Tertiary studies in software engineering (TS@SE) are widely used to synthesise evidence on a research topic systematically. As part of their protocol, TS@SE define inclusion and exclusion criteria (IC/EC) aimed at selecting those secondary studies (SS) to be included in the analysis. Aims: To provide a state of the art on the definition and application of IC/EC in TS@SE, and from the results of this analysis, we outline an emerging framework, TSICEC, to be used by SE researchers. Method: To provide the state of the art, we conducted a systematic mapping (SM) combining automatic search and snowballing over the body of SE scientific literature, which led to 50 papers after application of our own IC/EC. The extracted data was synthesised using content analysis. The results were used to define a first version of TSICEC. Results: The SM resulted in a coding schema, and a thorough analysis of the selected papers on the basis of this coding. Our TSICEC framework includes guidelines for the definition of IC/EC in TS@SE. Conclusion: This paper is a step forward establishing a foundation for researchers in two ways. As authors, understanding the different possibilities to define IC/EC and apply them to select SS. As readers, having an instrument to understand the methodological rigor upon which TS@SE may claim their findings., This work has been partially supported by the DOGO4ML Spanish research project (ref. PID2020-117191RB-I00)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Discovery of semantic non-syntactic joins

  • Maynou Yelamos, Marc
  • Nadal Francesch, Sergi
Data discovery is an essential step in the data integration pipeline involving finding datasets whose combined information provides relevant insights. Discovering joinable attributes requires assessing the closeness of the semantic concepts that two attributes represent, which is highly sensitive and dependent on the chosen similarity metric. The state of the art commonly approaches this task from a syntactic perspective, this is, performing comparisons based on the data values or on direct transformations (e.g., via hash functions). These approaches suffice when the two sets of instances share the same syntactic representation, but fail to detect cases in which the same semantic concept is represented by different sets of values, which we refer as semantic non-syntactic joins. This is a relevant problem in data lake scenarios, when the underlying datasets present high heterogeneity and lack of standardization. To that end, in this paper, we propose an empirical approach to detect semantic non-syntactic joins, which leverages, simultaneously, syntactic and semantic measurements of the data. We demonstrate that our approach is effective in detecting such kind of joins., Marc Maynou is supported by the EU’s Horizon Programme call, under Grant Agreements No. 101093164 (ExtremeXP), and Sergi Nadal is partially supported by the DOGO4ML project, funded by the Spanish Ministerio de Ciencia e Innovación under the funding scheme PID2020-117191RB-I00 / AEI / 10.13039/501100011033., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Insights on the use of software design principles in machine learning pipelines

  • López Cuesta, Lidia
  • Gómez Seoane, Cristina
  • Ayala Martínez, Claudia Patricia
The rapid expansion of Artificial Intelligence has driven a surge in the development of Machine Learning (ML) pipelines, essential for constructing and maintaining ML models. Despite the growing recognition of the importance of high-quality ML pipelines, there is limited actionable guidance on improving pipelines quality at the implementation level. The Object-Oriented (OO) paradigm and software design principles have been used in software engineering to enhance software system quality. In this context, this work explores the coverage of existing ML pipelines in relation to OO concepts and software design principles. Our findings reveal a limited use of OO mechanisms and minimal adherence to software design principles. Specifically, issues such as poor code organization and deficient levels of coupling and cohesion were observed. Our results can shed some light on the development of strategies to overcome these eficiencies for improving ML pipelines quality., This work was supported by the “Spanish Ministerio de Ciencia e Innovación”, under project PID2020-117191RB-I00, funding scheme AEI/https://doi.org/10.13039/ 501100011033., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

T-FREX: a transformer-based feature extraction method from mobile app reviews

  • Motger de la Encarnación, Joaquim
  • Miaschi, Alessio
  • Dell’Orletta, Felice
  • Franch Gutiérrez, Javier
  • Marco Gómez, Jordi
Mobile app reviews are a large-scale data source for software-related knowledge generation activities, including software maintenance, evolution and feedback analysis. Effective extraction of features (i.e., functionalities or characteristics) from these reviews is key to support analysis on the acceptance of these features, identification of relevant new feature requests and prioritization of feature development, among others. Traditional methods focus on syntactic pattern-based approaches, typically context-agnostic, evaluated on a closed set of apps, difficult to replicate and limited to a reduced set and domain of apps. Mean-while, the pervasiveness of Large Language Models (LLMs) based on the Transformer architecture in software engineering tasks lays the groundwork for empirical evaluation of the performance of these models to support feature extraction. In this study, we present T-FREX, a Transformer-based, fully automatic approach for mobile app review feature extraction. First, we collect a set of ground truth features from users in a real crowdsourced software recommendation platform and transfer them automatically into a dataset of app reviews. Then, we use this newly created dataset to fine-tune multiple LLMs on a named entity recognition task under different data configurations. We assess the performance of T- FREX with respect to this ground truth, and we complement our analysis by comparing T- FREX with a baseline method from the field. Finally, we assess the quality of new features predicted by T- FREX through an external human evaluation. Results show that T- FREX outperforms on average the traditional syntactic-based method, especially when discovering new features from a domain for which the model has been fine-tuned., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovacion under project I funding scheme PID2020-117191RB-100 I AEI/10.13039/501100011033. Alessio Miaschi and Felice Dell’ Orletta have also been supported by the PNRR project FAIR - Future AI Research (PE00000013), under the NRRP MUR program funded by the NextGenerationEU., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

There is no data science without data governance: a proposal based on knowledge graphs

  • Bilalli, Besim
  • Jovanovic, Petar
  • Nadal Francesch, Sergi
  • Queralt Calafat, Anna
  • Romero Moral, Óscar
Data Science and data-driven Artificial Intelligence are here to stay and they are expected to further transform the current global economy. From a technical point of view, there is an overall agreement that disciplines based on data require to combine data engineering and data analysis skills, but the fact is that data engineering is nowadays trailing and catching up with the rapid changes in the data analysis landscape. To unleash the real power of data, data-centric systems must be professionalized, i.e., operationalized and systematized, so that repetitive, time-consuming and error-prone tasks are automated. To such end, we propose our vision on next generation data governance for data-centric systems based on knowledge graphs. We claim that without the knowledge embedded in the data governance layer, Data Science will not unleash its potential., This work is supported by the Horizon Europe Programme under GA.101135513 (CYCLOPS) and GA. 101093164 (ExtremeXP) and the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RBI00 / AEI/10.13039/501100011033 (DOGO4ML)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Web API change-proneness prediction

  • Koçi, Rediana
  • Franch Gutiérrez, Javier
  • Jovanovic, Petar
  • Abelló Gamazo, Alberto
Change-proneness of software artifacts has been mainly related to the design characteristics and their previous history of changes. While these two aspects are essential and contribute significantly to the prediction, they leave out a critical factor: how the artifacts are being used. In the context of web APIs, consumers represent one of the main drivers of the change. Therefore, we propose a methodology for predicting the change-proneness of web API endpoint interfaces, taking into account not only design and change history but also their usage. Since the evolution of web APIs is and should be usage-driven, the way consumers use an API affects the future changes implemented by providers. Consequently, consumers' usage behavior contains essential information that contributes to identifying endpoints that are more prone to change. By considering the reasons behind changes, we introduce a set of metrics comprising design and usage aspects to be used as variables in prediction. To demonstrate the usefulness of the approach we perform an initial evaluation using a real-world web API. We quantify the introduced metrics using web API documentation, code, and usage logs in order to build a classifier able to predict with 82 % accuracy if an endpoint will change based on its design, history of changes, and usage characteristics., This paper has been funded by the Spanish Ministerio de Ciencia e Innovacion under project/funding scheme PID2020-117191RB-I00/AEI/10.13039/501100011033., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Towards continuous experiment-driven MLOps

  • Rajenthiram, Keerthiga
  • Abdullah, Milad
  • Gerostathopoulos, Ilias
  • Hnetynka, Petr
  • Bureš, Tomáš
  • Pons Recasens, Gerard
  • Bilalli, Besim
  • Queralt Calafat, Anna
Despite advancements in MLOps and AutoML, ML development still remains challenging for data scientists. First, there is poor support for and limited control over optimizing and evolving ML models. Second, there is lack of efficient mechanisms for continuous evolution of ML models which would leverage the knowledge gained in previous optimizations of the same or different models. We propose an experiment-driven MLOps approach which tackles these problems. Our approach relies on the concept of an experiment, which embodies a fully controllable optimization process. It introduces full traceability and repeatability to the optimization process, allows humans to be in full control of it, and enables continuous improvement of the ML system. Importantly, it also establishes knowledge, which is carried over and built across a series of experiments and allows for improving the efficiency of experimentation over time. We demonstrate our approach through its realization and application in the ExtremeXp11https://extremexp.eu/ project (Horizon Europe)., This work has been partially supported by the EU project ExtremeXP grant agreement 101093164, by Charles University institutional funding SVV 260698, and by the Spanish Ministerio de Ciencia e Innovación under project PID2020- 117191RB-I00/ AEI/10.13039/501100011033 (DOGO4ML)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Mobile feature-oriented knowledge base generation using knowledge graphs

  • Motger de la Encarnación, Joaquim
  • Franch Gutiérrez, Javier
  • Marco Gómez, Jordi
Knowledge bases are centralized repositories used for developing knowledge-oriented information systems. They are essential for adaptive, specialized knowledge in dialogue systems, supporting up-to-date domain-specific discussions with users. However, designing large-scale knowledge bases presents multiple challenges in data collection and knowledge exploitation. Knowledge graphs provide various research opportunities by integrating decentralized data and generating advanced knowledge. Our contribution presents a knowledge base in the form of a knowledge graph for extended knowledge generation for mobile apps and features extracted from related natural language documents. Our work encompasses the knowledge graph completion and deductive and inductive knowledge requests. We evaluated the effectiveness and performance of these knowledge strategies, which can be used as on-demand knowledge requests used by third-party software systems., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Goal-oriented models for teaching and understanding data structures

  • Franch Gutiérrez, Javier
  • Ruiz, Marcela
Most computer science curricula include a compulsory course on data structures. Students are prone to memorise facts about data structures instead of understanding the essence of underlying concepts. This can be explained by the fact that learning the basics of each data structure, the difference with each other, and the adequacy of each of them to the most appropriate context of use, is far from trivial. This paper explores the idea of providing adequate levels of abstractions to describe data structures from an intentional point of view. Our hypothesis is that adopting a goal-oriented perspective could emphasise the main goals of each data structure, its qualities, and its relationships with the potential context of use. Following this hypothesis, in this paper we present the use of iStar2.0 to teach and understand data structures. We conducted a comparative quasi-experiment with undergraduate students to evaluate the effectiveness of the approach. Significant results show the great potential of goal modeling for teaching technical courses like data structures. We conclude this paper by reflecting on further teaching and conceptual modeling research to be conducted in this field., This work has been partially supported by the by the DOGO4ML Spanish research project (ref. PID2020-117191RB-I00), the Digitalization Initiative of the Canton of Zü-rich (DIZH), and ZHAW Digital., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Leveraging requirements elicitation through software requirement patterns and LLMs

  • Franch Gutiérrez, Javier
  • Gnesi, Stefania
  • Paccosi, Federico
  • Quer, Carme
  • Semini, Laura
Software requirement patterns (SRPs) is one of the many techniques that contribute to requirements elicitation. At this respect, the emergence of large language models (LLMs) opens the door to cost-effective strategies to create and use SRPs. Still, the stochastic nature of LLMs threatens the inherent quality of requirements reuse and consequently, that of the elicitation process. [Question/problem] In this scientific evaluation paper, we investigate whether and how LLMs can be used in order to create an SRP catalogue and elicit requirements from it. [Principal ideas/results] SRPs can be effectively extracted by querying an LLM through appropriate prompts, but still expert assessment is key in order to deliver the best results. LLM-driven generation of questions to stakeholders for eliciting requirements from these SRPs is feasible but suffers from deficiencies such as excessive number of repetitions and out of scope requirements. [Contribution] We show that (1) LLMs can be embedded into the requirements elicitation process through a pattern instantiation-based strategy, but at the same time (2) the current state of LLM technologies requires expert assessment at a large extent., This paper has been partially funded by the Spanish Ministerio de Ciencia e Innovación under project/funding scheme PID2020-117191RBI00/AEI/10.13039/501100011033, by the Italian MIUR, under project PRIN 2022 STENDHAL and PNRR Project Securing sOftware Platforms., Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Hyppo: using equivalences to optimize pipelines in exploratory machine learning

  • Kontaxakis, Antonios
  • Sacharidis, Dimitris
  • Simitsis, Alkis
  • Abelló Gamazo, Alberto
  • Nadal Francesch, Sergi
We present HYPPO, a novel system to optimize pipelines encountered in exploratory machine learning. HYPPO exploits alternative computational paths of artifacts from past executions to derive better execution plans while reusing materialized artifacts. Adding alternative computations introduces new challenges for exploratory machine learning regarding workload representation, system architecture, and optimal execution plan generation. To this end, we present a novel workload representation based on directed hypergraphs, and we formulate the problem of discovering the optimal execution plan as a search problem over directed hypergraphs and that of selecting artifacts to materialize as an optimization problem. A thorough experimental evaluation shows that HYPPO results in plans that are typically one order (up to two orders) of magnitude faster and cheaper than the non-optimized pipeline and considerably (up to one order of magnitude) faster and cheaper than plans generated by the state of the art when materializing artifacts is possible. Lastly, our evaluation reveals that HYPPO reduces the cost by 3–4× even when materialization cannot be exploited., This work has been partially supported by the H2020-MSCAITN-2020 DEDS (GA.955895), the EU-HORIZON programmes FAIR-CORE4EOSC (GA.101057264), CREXDATA (GA.101092749), ExtremeXP (GA.101093164), and the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RB-I00/AEI/10.13039/501100011033 (DOGO4ML)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Software analytics tools: an intentional view

  • Franch Gutiérrez, Javier
Software analytic tools consume big amounts of data coming from either (or both) the software development process or the system usage and aggregate them into indicators which are rendered to different types of stakeholders, also offering them a portfolio of techniques and capabilities such as what-if analysis, prediction and alerts. Precisely, the variety of stakeholders and the different goals they pursue justifies the convenience of performing an intentional analysis of the use of software analytics tools. With this aim, we first enumerate the different stakeholders and identify their intentional relationships with software analytics tools in the form of dependencies. Then, we focus on one particular stakeholder, namely the requirements engineer, and identify further intentional elements represented in a strategic rationale model. The resulting model provides an abstract view of the domain which may help stakeholders when deciding on the adoption of software analytic tools in their particular context., This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Measuring discord among multidimensional data sources

  • Abelló Gamazo, Alberto
  • Cheney, James
Data integration is a classical problem in databases, typically decomposed into schema matching, entity matching and record merging. To solve the latter, it is mostly assumed that ground truth can be determined, either as master data or from user feedback. However, in many cases, this is not the case because firstly the merging processes cannot be accurate enough, and also the data gathering processes in the different sources are simply imperfect and cannot provide high quality data. Instead of enforcing consistency, we propose to evaluate how concordant or discordant sources are as a measure of trustworthiness (the more discordant are the sources, the less we can trust their data). Thus, we define the discord measurement problem in which given a set of uncertain raw observations or aggregate results (such as case/hospitalization/death data relevant to COVID-19) and information on the alignment of different data (for example, cases and deaths), we wish to assess whether the different sources are concordant, or if not, measure how discordant they are., The work of Alberto Abelló has been done under project PID2020- 117191RB-I00 funded by MCIN/ AEI /10.13039/501100011033. The work of James Cheney was supported by ERC Consolidator Grant Skye (grant number 682315)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

The knowledge graph lifecycle in NTT DATA

  • Flores Herrera, Javier de Jesús
  • Jamin, Emmanuel
  • Nadal Francesch, Sergi
  • Romero Moral, Óscar
The Semantic Business Unit (SEMBU) in NTT DATA aims to increase the semantic interoper ability and accessibility of European institutions’ data projects by following Linked Open Data (LOD) principles to build controlled vocabularies and produce Knowledge Graphs (KGs). One of its most notable projects revolves around the CORDIS portal1, which publishes information about research and innovation projects funded by the European Commission. SEMBU pursues two main goals: (i) expose semantic data related to CORDIS via a SPARQL endpoint that facilitates access and reuse of quality scientific-related data, and (ii) design an efficient, incremental, and automated KG lifecycle to be used as a reference in other data projects. To that end, we have adopted state-of-the-art semantic technologies to support the creation and management of the KG with the goal of centralizing knowledge and providing an overall view of data assets that improve data governance, maintenance, and external interaction by data consumers. We have also identified some of their limitations which are tackled via an industrial PhD. This paper reports our experience, the obstacles, and proposals for generating and maintaining the CORDIS KG., This work was partly funded by the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RBI00 (DOGO4ML). Javier Flores is supported by contract 2020-DI-027 of the Industrial Doctorate Program of the Government of Catalonia and CONACYT’s scholarship. Sergi Nadal is partly supported by the Spanish Ministerio de Ciencia e Innovación, as well as the European Union - NextGenerationEU, under project FJC2020-045809-I /AEI/10.13039/501100011033., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Architectural decisions in AI-based systems: an ontological view

  • Franch Gutiérrez, Javier
  • Martínez Fernández, Silverio Juan
  • Ayala Martínez, Claudia Patricia
  • Gómez Seoane, Cristina
Architecting AI-based systems entails making some decisions that are particular to this type of systems. Therefore, it becomes necessary to gather all necessary knowledge to inform such decisions, and to articulate this knowledge in a form that facilitates knowledge transfer among different AI projects. In this exploratory paper, we first present the main results of a literature survey in the area, and then we propose a preliminary ontology for architectural decision making, which we exemplify using a subset of the papers selected in the literature review. In the discussion, we remark on the variety of decision types and system contexts, highlighting the need to further investigate the current state of research and practice in this area. Besides, we summarize our plans to move along this research area by widening the literature review and incorporating more AI-related concepts to this first version of the ontology., This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Applying transfer learning to sentiment analysis in social media

  • de Arriba Serra, Ariadna
  • Oriol Hilari, Marc
  • Franch Gutiérrez, Javier
Context: Sentiment analysis is an NLP technique that can be used to automatically obtain the sentiment of a crowd of end-users regarding a software application. However, applying sentiment analysis is a difficult task, especially considering the need of obtaining enough good quality data for training a Machine Learning (ML) model. To address this challenge, transfer learning can help us save time and get better performance results with a limited amount of data. Objective: In this paper, we aim at identifying to which degree transfer learning improves the results of sentiment analysis of messages shared by end-users in social media. Method: We propose a tool-supported framework able to monitor and analyze the sentiment of tweets with different ML models and settings. Using the proposed framework, we apply transfer learning and conduct a set of experiments with multiple datasets. Results: The performance of different ML models with transfer learning from different datasets are obtained and discussed, showing how different factors affect the results, and discussing how they have to be considered when applying transfer learning., This work has been partially supported by the Spanish project DOGO4ML (contract PID2020-117191RB-I00)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Using metrics for code smells of ML pipelines

  • Costa, Dolors
  • Gómez Seoane, Cristina
  • Rey Juárez, Santiago del
  • Martínez Fernández, Silverio Juan
© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. https://dx.doi.org/0.1109/ICSA-C63560.2024.00055, ML pipelines, as key components of ML systems, shall be developed following quality assurance techniques. Unfortunately, it is often the case in which they present maintainability issues, due to the experimental nature of data collection and ML model construction. In this work, we perform a first evaluation of a set of metrics, proposed in previous research, for measuring the presence of code smells related to maintainability, in ML pipeline application examples. Moreover, we provide the lessons learnt and insights gained from this evaluation., This work has been funded by the National Spanish Program for Research Aimed at the Challenges of Society 2020. The project and funding identifiers are PID2020-117191RB-I00/ AEI/10.13039/501100011033., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

MApp-KG: Mobile app knowledge graph for document-based feature knowledge generation

  • Motger de la Encarnación, Joaquim
  • Franch Gutiérrez, Javier
  • Marco Gómez, Jordi
Mobile app repositories serve as large-scale crowdsourced information systems used for various document-based software engineering tasks, leveraging product descriptions, user reviews, and other natural language documents. Particularly, feature extraction (i.e., identifying functionalities or capabilities of a mobile app mentioned in these documents) is key for product recommendation, topic modelling, and feedback analysis. However, researchers often face domain-specific challenges in mining these repositories, including the integration of heterogeneous data sources, large-scale data collection, normalization and ground-truth generation for feature-oriented tasks. In this paper, we introduce MApp-KG, a combination of software resources and data artefacts in the field of mobile app repositories aimed at supporting feature-oriented knowledge generation tasks. Our contribution provides a framework for automatically constructing a knowledge graph that models a domain-specific catalog of natural language documents related to mobile applications. We distribute MApp-KG through a public triplestore, enabling its immediate use for future research and replication of our findings., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Data-driven agile requirements elicitation through the lenses of situational method engineering

  • Franch Gutiérrez, Javier
  • Henriksson, Aron
  • Ralyté, Jolita
  • Zdravkovic, Jelena
Ubiquitous digitalization has led to the continuous generation of large amounts of digital data, both in organizations and in society at large. In the requirements engineering community, there has been a growing interest in considering digital data as new sources for requirements elicitation, in addition to stake-holders. The volume, dynamics, and variety of data makes iterative requirements elicitation increasingly continuous, but also unstructured and complex, which current agile methods are unable to consider and manage in a systematic and efficient manner. There is also the need to support software evolution by enabling a synergy of stakeholder-driven requirements elicitation and management with data-driven approaches. In this study, we propose extension of agile requirements elicitation by applying situational method engineering. The research is grounded on two studies in the business domains of video games and online banking., The work presented in this paper is partially funded by the DOGO4ML Spanish research project, PID2020-117191RB-I00., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Performance analysis of distributed GPU-accelerated task-based workflows

  • Nogueira Lobo de Carvalho, Marcos
  • Queralt Calafat, Anna
  • Romero Moral, Óscar
  • Simitsis, Alkis
  • Tatu, Cristian
  • Badia Sala, Rosa Maria
We present an empirical approach to identify the key factors affecting the execution performance of task-based workflows on a High Performance Computing (HPC) infrastructure composed of heterogeneous CPU-GPU clusters. Our results reveal that the execution performance in distributed GPU-accelerated task-based workflows highly depends on several interrelated factors regarding the task algorithm, dataset, resources, and system employed. In addition, our analysis identifies key correlations among these factors, presents novel observations, and offers guidelines toward designing an automated method to handle task-based workflows in modern, high-compute capacity, CPU-GPU engines., This work has been partially supported by DEDS (H2020-MSCAITN2020) with grant agreement No. 955895, the EU-HORIZON programme CREXDATA under GA.101092749, the EU-HORIZON programme FAIR-CORE4EOSC under GA.101057264, the EUHORIZON programme EXTREMEXP under GA.101093164, the Spanish Government projects PID2019-107255GB and PID2020117191RB-I00/AEI/10.13039/501100011033andMCIN/AEI/10.13039 /501100011033 (CEX2021-001148-S), and by the Departament de Recerca i Universitats de la Generalitat de Catalunya (2021 SGR 00412, MPiEDist)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

A data-science pipeline to enable the interpretability of many-objective feature selection

  • Njoku, Uchechukwu Fortune
  • Abelló Gamazo, Alberto
  • Bilalli, Besim
  • Bontempi, Gianluca
Many-Objective Feature Selection (MOFS) approaches use four or more objectives to determine the relevance of a subset of features in a supervised learning task. As a consequence, MOFS typically returns a large set of non-dominated solutions, which have to be assessed by the data scientist in order to proceed with the final choice. Given the multi-variate nature of the assessment, which may include objectives (e.g., fairness) unrelated to predictive accuracy, this step is often not straightforward and suffers from the lack of existing tools. For instance, it is common to make use of a tabular presentation of the solutions, which provides little information about the trade-offs and the relationships between objectives over the set of solutions. Adopting a GA-based MOFS with six objectives (number of selected features, balanced accuracy, F1-Score, variance inflation factor, statistical parity, and equalised odds) for two feature selection tasks, this paper illustrates the complex challenge of assessing MOFS results and the need for a methodology to aid and justify the final choice of a solution., The project leading to this publication has received funding from the European Union’s Horizon 2020 research and innovation programme (G.A. No. 955895). A. Abelló and B. Bilalli are funded by the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RBI00, funding scheme AEI/10.13039/501100011033. G. Bontempi is supported by the Service Public de Wallonie Recherche under G.A. No. 2010235–ARIAC by DigitalWallonia4.ai. Computational resources have been provided by the Consortium des Équipements de Calcul Intensif, funded by the Fonds de la Recherche Scientifique de Belgique (G.A. No. 2.5020.11) and Walloon Region., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Unveiling competition dynamics in mobile app markets through user reviews

  • Motger de la Encarnación, Joaquim
  • Franch Gutiérrez, Javier
  • Gervasi, Vincenzo
  • Marco Gómez, Jordi
[Context and motivation] User reviews published in mobile app repositories are essential for understanding user satisfaction and engagement within a specific market segment. [Question/problem] Manual analysis of reviews is impractical due to the large data volume, and automated analysis faces challenges like data synthesis and reporting. This complicates the task for app providers in identifying patterns and significant events, especially in assessing the influence of competitor apps. Furthermore, review-based research is mostly limited to a single app or a single app provider, excluding potential competition analysis. Consequently, there is an open research challenge in leveraging user reviews to support cross-app analysis within a specific market segment. [Principal ideas/results] Following a case-study research method in the microblogging app market, we introduce an automatic, novel approach to support mobile app market analysis. Our approach leverages quantitative metrics and event detection techniques based on newly published user reviews. Significant events are proactively identified and summarized by comparing metric deviations with historical baseline indicators within the lifecycle of a mobile app. [Contribution] Results from our case study show empirical evidence of the detection of relevant events within the selected market segment, including software- or release-based events, contextual events and the emergence of new competitors., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project/funding scheme PID2020-117191RBI00/AEI/10.13039/501100011033., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Chatbots4Mobile: Feature-oriented knowledge base generation using natural language

  • Motger de la Encarnación, Joaquim
  • Franch Gutiérrez, Javier
  • Marco Gómez, Jordi
Chatbots4Mobile is a research project from the GESSI research group (UPC-BarcelonaTech) which aims at designing and developing a task oriented, knowledge based conversational agent to support mobile users in the process of managing and integrating the functionalities exposed by their own application portfolio. To support the design of the required knowledge base, the project focuses on the application of Natural Language Processing (NLP) techniques to infer extended knowledge about the features exposed by a subset of mobile applications, including feature extraction from app-related documents, syntactic and semantic similarity analysis between features, and intent/entity classification focused on functionality identification from user requests. As next steps, we are focusing on the evaluation of embedded linguistic knowledge in large language models, as well as the application of granular sentiment analysis techniques to discern biased documents and process sentiment-based user feedback., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Knowledge graphs for enhancing large language models in entity disambiguation

  • Pons Recasens, Gerard
  • Bilalli, Besim
  • Queralt Calafat, Anna
Recent advances in Large Language Models (LLMs) have positioned them as a prominent solution for Natural Language Processing tasks. Notably, they can approach these problems in a zero or few-shot manner, thereby eliminating the need for training or fine-tuning task-specific models. However, LLMs face some challenges, including hallucination and the presence of outdated knowledge or missing information from specific domains in the training data. These problems cannot be easily solved by retraining the models with new data as it is a time-consuming and expensive process. To mitigate these issues, Knowledge Graphs (KGs) have been proposed as a structured external source of information to enrich LLMs. With this idea, in this work we use KGs to enhance LLMs for zero-shot Entity Disambiguation (ED). For that purpose, we leverage the hierarchical representation of the entities’ classes in a KG to gradually prune the candidate space as well as the entities’ descriptions to enrich the input prompt with additional factual knowledge. Our evaluation on popular ED datasets shows that the proposed method outperforms non-enhanced and description-only enhanced LLMs, and has a higher degree of adaptability than task-specific models. Furthermore, we conduct an error analysis and discuss the impact of the leveraged KG’s semantic expressivity on the ED performance., This work is supported by the Horizon Europe Programme under GA.101093164 (ExtremeXP) and the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RB-I00/ AEI/10.13039/501100011033 (DOGO4ML). Anna Queralt is a Serra Húnter Fellow., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

TransFeatEx: a NLP pipeline for feature extraction

  • Gallego Marfa, Agustí
  • Motger de la Encarnación, Joaquim
  • Franch Gutiérrez, Javier
  • Marco Gómez, Jordi
Mobile app stores provide centralized access to a large data set of mobile app related natural language textual data, including developer’s documentation (e.g., descriptions, changelogs) and user-generated data (e.g., user reviews). Motivated by this context, multiple studies have focused on data-driven elicitation processes for the automatic extraction of the set of features exposed by a catalogue of applications and the inferred, extended knowledge that can be derived from this information. Moreover, with the emerging and generalization of large language models, traditional linguistic-based approaches can be significantly improved by the potential of the knowledge embedded in this kind of models. In this paper, we present TransFeatEx, a NLP-based feature extraction pipeline that combines the use of a RoBERTa-based model with the application of consolidated syntactic and semantic techniques. The pipeline is designed as a customizable, standalone service to be used either as a playground, experimentation tool or as a software component to be embedded into a third-party software system for batch-processing large document corpora. An example of a demo plan is showcased here: https://youtu.be/gfFyi_i_uTw, With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Merging datasets for emotion analysis

  • de Arriba Serra, Ariadna
  • Oriol Hilari, Marc
  • Franch Gutiérrez, Javier
Context. Applying sentiment analysis is in general a laborious task. Furthermore, if we add the task of getting a good quality dataset with balanced distribution and enough samples, the job becomes more complicated. Objective. We want to find out whether merging compatible datasets improves emotion analysis based on machine learning (ML) techniques, compared to the original, individual datasets. Method. We obtained two datasets with Covid-19-related tweets written in Spanish, and then built from them two new datasets combining the original ones with different consolidation of balance. We analyzed the results according to precision, recall, F1-score and accuracy. Results. The results obtained show that merging two datasets can improve the performance of ML models, particularly the F1-score, when the merging process follows a strategy that optimizes the balance of the resulting dataset. Conclusions. Merging two datasets can improve the performance of ML models for emotion analysis, whilst saving resources for labeling training data. This might be especially useful for several software engineering activities that leverage on ML-based emotion analysis techniques., This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Which design decisions in AI-enabled mobile applications contribute to greener AI?

  • Creus Castanyer, Roger
  • Martínez Fernández, Silverio Juan
  • Franch Gutiérrez, Javier
Background: The construction, evolution and usage of complex artificial intelligence (AI) models demand expensive computational resources. While currently available high-performance computing environments support well this complexity, the deployment of AI models in mobile devices, which is an increasing trend, is challenging. Mobile applications consist of environments with low computational resources and hence imply limitations in the design decisions during the AI-enabled software engineering lifecycle that balance the trade-off between the accuracy and the complexity of the mobile applications. Objective: Our objective is to systematically assess the trade-off between accuracy and complexity when deploying complex AI models (e.g. neural networks) to mobile devices, which have an implicit resource limitation. We aim to cover (i) the impact of the design decisions on the achievement of high-accuracy and low resource-consumption implementations; and (ii) the validation of profiling tools for systematically promoting greener AI. In this way, we aim to provide a quantitative analysis of the performance of AI-enabled applications in operation with respect to their design decisions. Method: This confirmatory registered report consists of a plan to conduct an empirical study to quantify the implications of the design decisions on AI-enabled applications performance and to report experiences of the end-to-end AI-enabled software engineering lifecycle. Concretely, we will implement both image-based and language-based neural networks in mobile applications to solve multiple image classification and text classification problems on different benchmark datasets. For that, we make use of the Android Studio and Unity3D frameworks for building mobile applications that work over Android; and TensorFlow Lite and Open Neural Network Exchange (ONNX) file formats for deploying the AI models in the applications. Overall, we plan to model the accuracy and complexity of AI-enabled applications in operation with respect to their design decisions and will provide tools for allowing practitioners to gain consciousness of the quantitative relationship between the design decisions and the green characteristics of study. Additionally, we will provide experiences in the end-to-end AI-enabled software lifecycle and discuss the challenges found, tools and practices for practitioners. Finally, we will provide an open-source data repository following the ESEM open science practices and containing all the experimentation, analysis and reports in our study., This work has been partially supported by the DOGO4ML Spanish research project (ref. PID2020-117191RB-I00) and by the the "Beatriz Galindo" Spanish Program BEAGAL18/00064., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

DOGO4ML: Development, operation and data governance for ML-based software systems

  • Ayala Martínez, Claudia Patricia
  • Bilalli, Besim
  • Gómez Seoane, Cristina
  • Martínez Fernández, Silverio Juan
Machine Learning based Software Systems (MLSS) are becoming increasingly pervasive in today’s society and can be found in virtually every domain. Building MLSS is challenging due to their interdisciplinary nature. MLSS engineering encompasses multiple disciplines, of which Data Engineering and Software Engineering appear as most relevant. The DOGO4ML project aims at reconciling these two disciplines for providing a holistic end-to-end framework to develop, operate and govern MLSS and their data. It proposes to combine and intertwine two software cycles: the DataOps and the DevOps lifecycles. The DataOps lifecycle manages the complexity of dealing with the big data needed by ML models, while the DevOps lifecycle is in charge of building the system that embeds these models. In this paper, we present the main vision and goals of the project as well as its expected contributions and outcomes. Although the project is in its initial stage, the progress of the research undertaken so far is detailed., This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Capturing analytical intents from text

  • Pons Recasens, Gerard
  • Dimic, Miona
  • Bilalli, Besim
The ability to extract valuable information from data is crucial for organizations and individuals who want to remain competitive in a constantly evolving data-driven environment. However, some of them lack the skills required to appropriately leverage the existing data analytics tools and methods. This problem is aggravated when the users are domain-experts but completely unfamiliar with data analytics terminology, as existing assistant tools, such as AutoML or Intelligent Discovery Assistants, require them to state their analytical intent (i.e., the type of data analysis they want to perform). To address this problem, we propose to capture the underlying analytical intent from textual problem descriptions by leveraging Large Language Models (LLMs). To this end, we propose a hierarchical categorization of analytical intents, along with a data collection methodology to obtain analytical problem descriptions for all of them in order to validate different approaches that aim to extract such intents from text. Next, we compare the performance of state-of-the-art approaches with LLMs, and then study the performance of different LLMs based on their characteristics and the impact of the source of validation data. Finally, we develop a prototype to showcase how our method could interact with existing AutoML systems., Gerard Pons is supported by the EU’s Horizon Programme call, under Grant Agreement No. 101093164 (ExtremeXP), and Besim Bilalli is partially supported by the DOGO4ML project, funded by the Spanish Ministerio de Ciencia i Innovación under the funding scheme PID2020-117191RB-I00/AEI/10.13039/501100011033., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

RE-Miner 2.0: a holistic framework for mining mobile application reviews

  • Tiessler Aguirre, Max
  • Motger de la Encarnación, Joaquim
In the domain of application stores and marketplaces, user reviews are crucial for supporting multiple requirements engineering tasks. Feature extraction, emotion classification, topic analysis, review type identification, and polarity analysis are key components in requirements prioritization, feedback gathering, and release planning. Empirical evaluation of these techniques is challenging due to data collection complexities and a lack of reproducible methods and available tools. Furthermore, existing studies often focus on isolated tasks, hindering a comprehensive analysis of user perceptions. This paper introduces RE-Miner 2.0, a work-in-progress tool that integrates multiple data extraction and analysis methods in a distributed environment (RE-Miner Ecosystem), enabling a multidimensional and detailed analysis of user feedback. It offers a web-based service for task integration and comparison, supported by persistent storage and a web application that allows analytical visualization of reviews. As a result, RE-Miner 2.0 provides a platform for task integration, replication, and comparison of review mining techniques. Bringing advancements in deep review analysis for requirements engineering. A demo of the tool is showcased here: https://www.youtube.com/watch?v=a11bHSCYqqM., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Evaluating quality of disparate data sources: A discord-driven approach

  • Akter, Yeasmin Ara
  • Abelló Gamazo, Alberto
  • Jovanovic, Petar
  • Sagi, Tomer
  • Hose, Katja
Among other measures of data quality, determining the reliability of conflicting values from different sources is especially challenging. Traditional data fusion approaches often infer correct values in simple cases, but struggle to handle variations in data granularity (such as differences in temporal, spatial, or categorical aggregations) and offer limited insight into the nature of disagreements. Thus, we propose a new source evaluation approach for numerical attributes that measures discordance (i.e., the extent to which sources differ from each other). Unlike existing methods that focus solely on point estimation, we allow both fine-grained and coarse-grained analysis, allowing more sophisticated data quality assessments. We employ a linear programming solver that transparently adapts to any data alignment expressed in a set of operators resembling relational algebra. Extensive experiments on real-world datasets demonstrate that our method generalizes existing truth discovery techniques measuring differences with Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and can adapt to diverse and complex scenarios., Y. A. Akter is funded by the EC Horizon 2020 research and innovation programme (DEDS: grant agreement No 955895). A. Abelló and P. Jovanovic are funded by the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RB-I00/AEI/10.13039/501100011033 (DOGO4ML) and the EC Horizon Europe programme (ExtremeXP: grant agreement No 101093164)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

How and why false denial constraints are discovered

  • Martín Garcia, Albert
  • Cunha de Almeida, Eduardo
  • Romero Moral, Óscar
  • Queralt Calafat, Anna
Denial Constraints (DCs) are a flexible formalism to express many types of data rules, making them a widely adopted tool for many applications. This flexibility led to the development of numerous algorithms to automatically discover DCs directly from data. However, few studies have been conducted on the quality of the discovered DCs. We experimentally quantify the lack of quality in the results obtained by state-of-the-art algorithms, showing how the proportion of discovered DCs that are false is rarely below 95%. We hypothesize that the common source of these erroneous DCs stems from the adoption of the current DC validity definition. We use a statistical approach to explain the mechanism leading to these results, and propose a redefinition of DC validity properties to avoid the acceptance of false DCs. We validate this redefinition experimentally, showing that it exclusively accepts true constraints of the data, and is reliable enough to discover DCs missed by domain experts. Additionally, we provide curated sets of golden DCs for each dataset used in our study, those generated by domain experts and those discovered using our approach., This work is supported by the Horizon Europe Programme under GA.101135513 (CyclOps) and the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RB-I00 / AEI/10.13039/ 501100011033 (DOGO4ML). Anna Queralt is a Serra-Húnter fellow. E. Almeida is funded by the CNPQ grants 302909/2022-2 and 444192/2024-7. Albert Martin is funded by the predoctoral program AGAUR-FI grants (2025 FI-1 00967) Joan Oró, which is backed by the Secretariat of Universities and Research of the Department of Research and Universities of the Generalitat of Catalonia, as well as the European Social Plus Fund., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Mining app reviews for user feedback analysis in requirements engineering: a project report

  • Motger de la Encarnación, Joaquim
  • Oriol Hilari, Marc
  • Tiessler Aguirre, Max
  • Franch Gutiérrez, Javier
  • Marco Gómez, Jordi
Mining app reviews has emerged as a valuable practice in requirements engineering, providing insights into feature usage trends, user satisfaction, and emerging software issues. While recent advances in natural language processing have enhanced review analysis, challenges persist in feature extraction, sentiment ambiguity, and the scalability of automated methods, among others. This project report presents our research efforts in app review mining, focusing on methodological, software-based, and data-driven contributions. We explore both supervised and unsupervised learning approaches, leveraging large language models for key tasks such as feature identification, competition analysis, and emotion extraction. Additionally, we develop open-source tools and datasets to support reproducibility and adoption of our methods. Our findings highlight the potential of large language models in automating user feedback analysis while identifying gaps that require further research, particularly in addressing model reliability and evaluation challenges., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Metrics for code smells of ML pipelines

  • Costa, Dolors
  • Gómez Seoane, Cristina
  • Martínez Fernández, Silverio Juan
ML pipelines, as key components of ML systems, shall be developed following quality assurance techniques. Unfortunately, it is often the case in which they present maintainability issues, due to the experimentatal nature of data collection and ML model construction. To address this problem, this work in progress proposes initial metrics to measure the presence of code smells in ML pipelines. These metrics reflect good software engineering practices for code quality of ML pipelines., This work was supported by the “Spanish Ministerio de Ciencia e Innovación” under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Energy efficiency of training neural network architectures: an empirical study

  • Xu, Yinlena
  • Martínez Fernández, Silverio Juan
  • Martínez Martínez, Matías-Sebastián
  • Franch Gutiérrez, Javier
The evaluation of Deep Learning (DL) models has traditionally focused on criteria such as accuracy, F1 score, and related measures. The increasing availability of high computational power environments allows the creation of deeper and more complex models. However, the computations needed to train such models entail a large carbon footprint. In this work, we study the relations between DL model architectures and their environmental impact in terms of energy consumed and CO2 emissions produced during training by means of an empirical study using Deep Convolutional Neural Networks. Concretely, we study: (i) the impact of the architecture and the location where the computations are hosted on the energy consumption and emissions produced; (ii) the trade-off between accuracy and energy efficiency; and (iii) the difference on the method of measurement of the energy consumed using software-based and hardware-based tools., This work has been supported by the Spanish project PID2020-117191RB-I00 funded by MCIN/AEI/10.13039/501100011033., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

RE-Miner: Mining mobile user reviews with feature extraction and emotion classification

  • Motger de la Encarnación, Joaquim
  • Tiessler Aguirre, Max
  • Oriol Hilari, Marc
  • Bertolín Rico, Irene
In the context of app stores, user reviews are pivotal on supporting multiple requirements engineering tasks. Among these, feature extraction and emotion classification play a crucial role in requirements prioritization, feedback gathering and release planning. Empirical evaluation of these techniques is impeded by data collection complexities and a lack of reproducible methods and available tools. Furthermore, existing studies often focus on isolated tasks, hindering a comprehensive analysis of user perceptions. This paper introduces RE-Miner, a work-in-progress tool integrating multiple feature extraction and emotion classification innovative methods, enabling a detailed analysis of feature-oriented user feedback. RE-Miner comprises a web-based service for task integration and comparison, and a web application for persistent storage and analytical visualization of reviews. As a result, RE-Miner provides a platform for seamless integration, replication, and comparison of review mining techniques, fostering advancements in feature extraction and emotion classification understanding for requirements engineering. A demo of the tool is showcased here: https://youtu.be/PFNCbborPuU., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Multi-agent debate strategies to enhance requirements engineering with large language models

  • Oriol Hilari, Marc
  • Motger de la Encarnación, Joaquim
  • Marco Gómez, Jordi
  • Franch Gutiérrez, Javier
Context: Large Language Model (LLM) agents are becoming widely used for various Requirements Engineering (RE) tasks. Research on improving their accuracy mainly focuses on prompt engineering, model fine-tuning, and retrieval augmented generation. However, these methods often treat models as isolated black boxes - relying on single-pass outputs without iterative refinement or collaboration, limiting robustness and adaptability. Objective: We propose that, just as human debates enhance accuracy and reduce bias in RE tasks by incorporating diverse perspectives, different LLM agents debating and collaborating may achieve similar improvements. Our goal is to investigate whether Multi-Agent Debate (MAD) strategies can enhance RE performance. Method: We conducted a systematic study of existing MAD strategies across various domains to identify their key characteristics. To assess their applicability in RE, we implemented and tested a preliminary MAD-based framework for RE classification. Results: Our study identified and categorized several MAD strategies, leading to a taxonomy outlining their core attributes. Our preliminary evaluation demonstrated the feasibility of applying MAD to RE classification. Conclusions: MAD presents a promising approach for improving LLM accuracy in RE tasks. This study provides a foundational understanding of MAD strategies, offering insights for future research and refinements in RE applications., This work has been supported by funding from the HIVEMIND project – Horizon Europe call HORIZON-CL4-2024- DIGITAL-EMERGING-01 under Grant Agreement Number 101189745; and by the Spanish Ministerio de Ciencia e Innovación under project funding scheme PID2020-117191RB-I00/ AEI/10.13039/501100011033., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Hyppo: efficient discovery and execution of data science pipelines in collaborative environments

  • Kontaxakis, Antonios
  • Sacharidis, Dimitris
  • Simitsis, Alkis
  • Abelló Gamazo, Alberto
  • Nadal Francesch, Sergi
In exploratory data science and machine learning (ML), developing an effective and efficient solution involves the exploration of numerous pipelines per dataset, considering various combinations of data preprocessing, feature selection, model selection, evaluation metrics, etc. This is an iterative process of trial and error, where a pipeline is revised and refined until a satisfactory level of performance is achieved. Such a process can be substantially improved by leveraging collaboration opportunities and automated pipeline optimization. We demonstrate Hyppo, a scalable collaboration system designed for interactive discovery and automated optimization, resulting in a system that supports evaluating and sharing pipelines with minimal overhead. Hyppo takes advantage of numerous optimization possibilities as exploiting equivalences among tasks, sharing computations, artifact materialization, and reuse from past executions to derive better execution plans. We showcase Hyppo with open data and pipelines from popular public code repositories and catalogs. Demo video: https://youtu.be/iilPV_xtXwM, This work has been partially supported by the H2020-MSCA-ITN-2020 DEDS (GA.955895), the European Union’s Horizon Europe Research and Innovation programmes DataGEMS (GA.101188 416), CREXDATA (GA.101092749), ExtremeXP (GA.101093164), and the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191 RB-I00/AEI/10.13039/50 1100011033 (DOGO4ML)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Adaptive task-oriented chatbots using feature-based knowledge bases

  • Campàs Gené, Carla
  • Motger de la Encarnación, Joaquim
  • Franch Gutiérrez, Javier
  • Marco Gómez, Jordi
Task-oriented chatbots relying on a knowledge base for domain-specific content exploitation have been largely addressed in research and industry applications. Despite this, multiple challenges remain to be fully conquered, including adaptive knowledge mechanisms, personalization for user-specific demands, and composite intent resolution. To address these challenges, in this paper, we present a work-in-progress summary of a task-oriented, knowledge-based chatbot in the field of mobile software ecosystems. The chatbot is designed to assist users in the combined use of multiple features from different applications. The proposed knowledge base and the machine learning pipeline supporting the chatbot technical core are designed to: (i) effectively use user context, (ii) process runtime feedback, (iii) use user historical data, and (iv) automatically infer slot values and dependent actions. With this report, we expect to lay the groundwork for future development stages and user validation studies., With the support from the Secretariat for Universities and Research of the Ministry of Business and Knowledge of the Government of Catalonia and the European Social Fund. This paper has been funded by the Spanish Ministerio de Ciencia e Innovación under project / funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Mitigating data sparsity in integrated data through text conceptualization

  • Rahman, Md Ataur
  • Nadal Francesch, Sergi
  • Romero Moral, Óscar
  • Sacharidis, Dimitris
We study the data sparsity problem for data generated from an integration system. We approach the problem from a textual information extraction perspective and propose to conceptualize external documents using the concepts in the integrated schema. We present THOR, a novel system that, unlike related approaches, neither relies on complex rules nor models trained with large annotated corpus, but on the integrated data and its schema without the need for human annotations. An extensive evaluation on the text conceptualization task demonstrates the superiority of our approach in terms of F1-score, effort and use of resources over the state-of-the-art language models., Md Ataur Rahman is funded by the EC Horizon 2020 research and innovation programme (DEDS: grant agreement No 955895). Sergi Nadal and Oscar Romero are funded by the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RB-I00/AEI/10.13039/501100011033 (DOGO4ML) and the EC Horizon Europe programme (ExtremeXP: grant agreement No 101093164)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

A data quality framework for graph-based virtual data integration systems

  • Li, Yalei
  • Nadal Francesch, Sergi
  • Romero Moral, Óscar
Data Quality (DQ) plays a critical role in data integration. Up to now, DQ has mostly been addressed from a single database perspective. Popular DQ frameworks rely on Integrity Constraints (IC) to enforce valid application semantics, which lead to the Denial Constraint (DC) formalism which models a broad range of ICs in real-world applications. Yet, current approaches are rather monolithic, considering a single database and do not suit data integration scenarios. In this paper, we address DQ for data integration systems. Specifically, we extend virtual data integration systems to elicit DCs from disparate data sources to be integrated, using DC-related state-of-the-art, and propagate them to the integrated schema (global DCs). Then, we propose a method to manage global DCs and identify (i) minimal DCs and (ii) potential clashes between them., This work was partly supported by the DOGO4ML project, funded by the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RB-I00. Sergi Nadal is partly supported by the Spanish Ministerio de Ciencia e Innovación, as well as the European Union - NextGenerationEU, under project FJC2020-045809-I., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

HealthMesh: An architectural framework for federated healthcare data management

  • Bisquert Parés, Aniol
  • Hmimou Ham Man, Achraf
  • Berral García, Josep Lluís
  • Gutiérrez Torre, Alberto
  • Romero Moral, Óscar
Recently, significant milestones have been achieved in the field of healthcare data analysis. However, alongside these accomplishments, substantial data-related challenges have emerged in the domain of big data management. Modern healthcare projects are no more dealing with a single data repository but many heterogeneous ones and must overcome data variety, privacy and governance issues. Yet, current solutions face a privacy-decentralization trade-off. To address this dual challenge, we introduce HealthMesh, a novel layered architectural framework based on the Data Mesh principles, providing a domaindecentralised paradigm. In addition, the framework incorporates a Semantic Data Model which establishes robust governance, enables interoperability and guarantees policy compliance for all the data assets. To demonstrate the capabilities of the proposed approach, we provide an illustrative example inspired by the use case of the INCISIVE project for breast cancer analytics. Overall, this work makes a significant contribution on collecting key challenges, identifying actors and providing a set of components and guidelines for establishing a holistic framework for the complex field of healthcare data management., The work reported here was supported by the EU-H2020 programme under GA.952179 INCISIVE, Horizon Europe Programme under GA.101095717 (SECURED) and GA.101135513 (CYCLOPS), the Spanish Research State Agency (MICINN/AEI, ERDF/FEDER) under GA.MCIN/AEI/10.13039/501100011033/FEDER (UE DALEST), the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RB-I00 / AEI/10.13039/501100011033 (DOGO4ML), and the Generalitat de Catalunya (AGAUR) 2021-SGR-00478 (CROMAI)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Impact study of NoSQL refactoring in SkyServer database

  • Gallinucci, Enrico
  • Golfarelli, Matteo
  • Radwan, Wafaa
  • Zarate Calderón, Gabriel
  • Abelló Gamazo, Alberto
Data modeling in NoSQL databases is notoriously complex and driven by multiple and possibly conflicting requirements. Researchers have proposed methodologies to optimize schema design of a given domain for a given workload; however, due to the agile environment in which NoSQL databases are usually employed, both domain and workload are frequently subject to changes and evolution- possibly neutralizing the benefits of optimization. When this happens, the benefits of a new optimal schema design must be weighed against the costs of migrating the data. In this work, we empirically show the benefits of schema redesign in a real publicly available database. In particular, we identify multiple snapshots (in terms of domain extension and querying workload) in the 20+ years evolution of SkyServer, demonstrate how NoSQL schema optimization at a given time can later backfire, and evaluate the conditions under which data migration becomes beneficial. This takes us to define the foundations and challenges of a framework for continuous NoSQL database refactoring, with the goal of helping DBAs and data engineers decide if, when, and how a NoSQL database should be reconsidered to restore schema design optimality., This work has been partially supported by the Spanish Ministerio de Ciencia e Innovación under project PID2020-117191RB-I00/AEI/10.13039/501100011033 (DOGO4ML)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

What about emotions? Guiding fine-grained emotion extraction from mobile app reviews

  • Motger de la Encarnación, Joaquim
  • Oriol Hilari, Marc
  • Tiessler Aguirre, Max
  • Franch Gutiérrez, Javier
  • Marco Gómez, Jordi
Opinion mining plays a vital role in analysing user feedback and extracting insights from textual data. While most research focuses on sentiment polarity (e.g., positive, negative, neutral), fine-grained emotion classification in app reviews remains underexplored. Fine-grained emotion classification is thus needed to better understand users’ affective responses and support downstream tasks such as feature-emotion analysis, user-oriented release planning, and issue triaging. This paper addresses this gap by identifying and addressing the challenges and limitations in fine-grained emotion analysis in the context of app reviews. Our study adapts Plutchik’s emotion taxonomy to app reviews by developing a structured annotation framework and dataset. Through an iterative human annotation process, we define clear annotation guidelines and document key challenges in emotion classification. Additionally, we evaluate the feasibility of automating emotion annotation using large language models, assessing their cost-effectiveness and agreement with human-labelled data. Our findings reveal that while large language models significantly reduce manual effort and maintain substantial agreement with human annotators, full automation remains challenging due to the complexity of emotional interpretation. This work contributes to opinion mining in requirements engineering by providing structured guidelines, an annotated dataset, and insights for developing automated pipelines to capture the complexity of emotions in app reviews., This work has been supported by funding from the HIVEMIND project – Horizon Europe call HORIZON-CL4-2024- DIGITAL-EMERGING-01 under Grant Agreement Number 101189745. This paper has been funded by the Spanish Ministerio de Ciencia e Innovacióin under project/funding scheme PID2020-117191RB-I00 / AEI/10.13039/501100011033., Peer Reviewed, Postprint (author's final draft)