Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 16
Encontrada(s) 1 página(s)

Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Big data and machine learning to improve european grapevine moth (Lobesia botrana) predictions

  • Balduque Gil, J.
  • Lacueva Pérez, F. J.
  • Labata Lezaun, G.
  • Del Hoyo Alonso, R.
  • Ilarri, S.
  • Sánchez Hernández, E.
  • Martín Ramos, P.
  • Barriuso Vargas, J. J.
Machine Learning (ML) techniques can be used to convert Big Data into valuable information for agri-environmental applications, such as predictive pest modeling. Lobesia botrana (Denis & Schiffermüller) 1775 (Lepidoptera: Tortricidae) is one of the main pests of grapevine, causing high productivity losses in some vineyards worldwide. This work focuses on the optimization of the Touzeau model, a classical correlation model between temperature and L. botrana development using data-driven models. Data collected from field observations were combined with 30 GB of registered weather data updated every 30 min to train the ML models and make predictions on this pest’s flights, as well as to assess the accuracy of both Touzeau and ML models. The results obtained highlight a much higher F1 score of the ML models in comparison with the Touzeau model. The best-performing model was an artificial neural network of four layers, which considered several variables together and not only the temperature, taking advantage of the ability of ML models to find relationships in nonlinear systems. Despite the room for improvement of artificial intelligence-based models, the process and results presented herein highlight the benefits of ML applied to agricultural pest management strategies.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

I am all EARS: Using open data and knowledge graph embeddings for music recommendations

  • Bertram, Niels
  • Dunkel, Jürgen
  • Hermoso, Ramón
Music streaming platforms offer music listeners an overwhelming choice of music. Therefore, users of streaming platforms need the support of music recommendation systems to find music that suits their personal taste. Currently, a new class of recommender systems based on knowledge graph embeddings promises to improve the quality of recommendations, in particular to provide diverse and novel recommendations. This paper investigates how knowledge graph embeddings can improve music recommendations. First, it is shown how a collaborative knowledge graph can be derived from open music data sources. Based on this knowledge graph, the music recommender system EARS (knowledge graph Embedding-based Artist Recommender System) is presented in detail, with particular emphasis on recommendation diversity and explainability. Finally, a comprehensive evaluation with real-world data is conducted, comparing of different embeddings and investigating the influence of different types of knowledge.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Enriched traffic datasets for the city of Madrid: Integrating data from traffic sensors, the road infrastructure, calendar data and weather data

  • Gómez, Iván
  • Ilarri, Sergio
The proliferation of urban areas and the concurrent increase in vehicular mobility have escalated the urgency for advanced traffic management solutions. This data article introduces two traffic datasets from Madrid, collected between June 2022 and February 2024, to address the challenges of traffic management in urban areas. The first dataset provides detailed traffic flow measurements (vehicles per hour) from urban sensors and road networks, enriched with weather data, calendar data and road infrastructure details from OpenStreetMap. This combination allows for an in-depth analysis of urban mobility. Through preprocessing, data quality is ensured by eliminating inconsistent sensor readings. The second dataset is enhanced for advanced predictive modelling. It includes time-based transformations and a tailored preprocessing pipeline that standardizes numeric data, applies one-hot encoding to categorical features, and uses ordinal encoding for specific features. In constructing the datasets, we initially employed the k-means algorithm to cluster data from multiple sensors, thereby highlighting the most representative ones. This clustering can be adapted or modified according to the user's needs, ensuring flexibility for various analyses and applications. This work underscores the importance of advanced datasets in urban planning and highlights the versatility of these resources for multiple practical applications. We highlight the relevance of the collected data for a variety of essential purposes, including traffic prediction, infrastructure planning, studies on the environmental impact of traffic, event planning, and conducting simulations. These datasets not only provide a solid foundation for academic research but also for designing and implementing more effective and sustainable traffic policies. Furthermore, all related datasets, source code, and documentation have been made publicly available, encouraging further research and practical applications in traffic management and urban planning.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

PDE-LDDMM meets NODEs: Introducing neural ordinary differential equation solvers in PDE-constrained Large Deformation Diffeomorphic Metric Mapping

  • Hernandez, Monica
Non-rigid image registration is a crucial task in various medical applications, allowing the alignment of images with complex spatial or temporal variations. This paper introduces NODEO-LDDMM and NODEO-PDE-LDDMM, two innovative deep-learning-based approaches that bridge the gap between Large Deformation Diffeomorphic Metric Mapping (LDDMM) and neural ordinary differential equations (NODEs). LDDMM and PDE-LDDMM offer mathematically well-established formulations for diffeomorphic registration, while NODEs provide the flexibility of deep-learning in the solution of the ODEs involved in both methods. Both NODEO-LDDMM and NODEO-PDE-LDDMM include the strengths of deep-learning into LDDMM, enabling a robust optimization with a good balance between accuracy and transformation smoothness in their solutions. Our proposed methods reached or outperformed their traditional counterparts and the nearly diffeomorphic deep-learning-based approaches selected as benchmarks. This work contributes to advancing non-rigid image registration techniques, with a methodology suited to overcome some of the limitations of deep-learning in medical image registration.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Developing machine learning models from multisourced real-world datasets to enhance smart-farming practices

  • Lacueva-Pérez, Francisco José
  • Hoyo-Alonso, Rafael del
  • Labata-Leazaún, Gorka
  • Barriuso-Vargas, Juan José
  • Ilarri-Artigas, Sergio
Smart farming technologies empower farmers to achieve sustainability by enabling data-driven decision-making. For instance, smart farming can contribute to farm sustainability through optimized pest control. By utilizing pest risk prediction models, farmers can conserve resources and minimize environmental impact by avoiding unnecessary treatments. However, effective crop pest control relies on timely treatments at specific phenological stages. Therefore, the ability to accurately predict phenological development becomes a crucial factor in increasing farm sustainability. This paper describes the design, development and evaluation of a set of Machine Learning models that predict the phenology of grapevines. The models were trained on multisourced data that combine 9 different datasets with different temporal and spatial resolutions. The authors evaluated and compared different machine learning algorithms to predict 9 different phenological stages of grapevines. The models that performed best also included data derived from Sentinel-2 images, which suggests that multispectral satellite images could be used to monitor and predict woody plant phenology. A key contribution of our proposal is the combination of multiple data sources and a fine-grained prediction aimed at distinguishing among 9 phenological states.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Addressing cross-population domain shift in chest X-ray classification through supervised adversarial domain adaptation

  • Musa, Aminu
  • Prasad, Rajesh
  • Hernández, Mónica
Medical image analysis, empowered by artificial intelligence (AI), plays a crucial role in modern healthcare diagnostics. However, the effectiveness of machine learning models hinges on their ability to generalize to diverse patient populations, presenting domain shift challenges. This study explores the domain shift problem in chest X-ray classification, focusing on cross-population variations, especially in underrepresented groups. We analyze the impact of domain shifts across three population datasets acting as sources using a Nigerian chest X-ray dataset acting as the target. Model performance is evaluated to assess disparities between source and target populations, revealing large discrepancies when the models trained on a source were applied to the target domain. To address with the evident domain shift among the populations, we propose a supervised adversarial domain adaptation (ADA) technique. The feature extractor is first trained on the source domain using a supervised loss function in ADA. The feature extractor is then frozen, and an adversarial domain discriminator is introduced to distinguish between the source and target domains. Adversarial training fine-tunes the feature extractor, making features from both domains indistinguishable, thereby creating domain-invariant features. The technique was evaluated on the Nigerian dataset, showing significant improvements in chest X-ray classification performance. The proposed model achieved a 90.08% accuracy and a 96% AUC score, outperforming existing approaches such as multi-task learning (MTL) and continual learning (CL). This research highlights the importance of developing domain-aware models in AI-driven healthcare, offering a solution to cross-population domain shift challenges in medical imaging.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

When Good Enough is the Best Option: Use of Digital Sufficiency to Fight Climate Change

  • Tirel, Nicolas
  • Ilarri, Sergio
  • Roose, Philippe
  • Noureddine, Adel
  • Le Goaër, Olivier
The ICT sector has seen many improvements in productivity and efficiency in the past decades. However, its total energy consumed, as well as its carbon emissions, has never decreased. This is mostly due to a systemic phenomenon called the rebound effect: although each process consumes less energy, the global increase in process usage results in an unexpected growth in carbon emissions. In this paper, we present energy-efficient techniques and we argue why it is necessary to combine efficiency with a new approach. This novel approach is digital sufficiency, involving users and developers to reduce the demand in addition to efficiency improvements.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Wind of change for agent decisions and innovation diffusion: The ASPID predictive model for technology adoption

  • Sáenz-Royo, Carlos
  • Hermoso, Ramón
  • Chiclana, Francisco
Innovations are part of human evolution and are essential for survival. However, traditional innovation diffusion models do not contemplate the possibility of innovation failure and focus on imitation social processes that require historical data for their estimation, providing only ex-post information, which limits their usefulness for risk management operations. This paper proposes a more general new model (ASPID: AbS-based Predictive Innovation Diffusion model) focusing on the decisions of agents who exhibit intentional limited rationality (IBR). ASPID provides: i) a greater depth of ex-post analysis than the classical models; ii) ex-ante information on the innovation diffusion process based on the characteristics of the target agents, the quality of the innovation, and the network topology of their relationships; iii) the success probability since innovations can fail. The model’s versatility allows it to adapt to any information level, from the most aggregated to the most detailed. Some exante and ex-post examples are presented to support the contribution.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Energy-aware software design: A hardware and behavioural consumption scoring and labelling algorithm

  • Larracoechea, Jorge Andrés
  • Roose, Philippe
  • Ilarri, Sergio
Energy labelling of electric appliances, among other goods and services, matters to both consumers and producers. It aids the consumers in gathering an expectation of how much electricity the good they are acquiring will consume, and it establishes guidelines that producers can follow to create more efficient products and services. Energy labelling is, nevertheless, nearly absent for software due to the heterogeneous hardware execution environments and unequal circumstances. Throughout this article, we provide an equal ground for software energy labelling by providing an intelligent prospective behavioural, hardware and energy-consumption patterns rating algorithm meant for the design phase of the Software-Development Life Cycle, as opposed to the existing approaches whose labelling procedure applies to the implementation phase onward. We expect our categorization method to guide software developers, helping them to decrease their energy consumption by raising their awareness of the energy footprint that their design choices produce; culminating in aware consumers who know what energy usage to expect from the software they consume.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Advanced prediction of traffic at different temporal scales using heterogeneous data sources

  • Gómez, Iván
  • Ilarri, Sergio
Efficient urban traffic management is a crucial challenge in modern smart cities, especially in densely populated areas with complex and dynamic traffic conditions. In this paper, we tackle the traffic prediction problem and present a lightweight architecture that combines sensor embeddings with dense layers, sustaining strong performance across both short- and long-term forecasting horizons while substantially reducing training time and enabling fast inference times. In comparative evaluations, our approach matches or surpasses the accuracy of more complex methods and consistently improves efficiency. To foster reproducibility, we release the code along with an enriched dataset that integrates traffic flows with contextual features such as weather conditions, temporal variables, and urban attributes. The richness and coverage of this dataset exceed those of existing public resources, enabling deeper and more comprehensive analyses of traffic dynamics. Overall, we demonstrate that a lightweight, well-designed architecture can achieve high performance and practical scalability for urban mobility management.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Innovaciones en la Gestión del Tráfico Urbano Utilizando Fuentes de Datos Heterogéneas e IA Un Enfoque Integrado

  • Gómez Pascual, Iván
  • Ilarri Artigas, Sergio
En este breve artículo, presentamos nuestra línea de investigación actual y futura centrada en la mejora de las técnicas de gestión y predicción del tráfico urbano. Nuestra metodología se basa en la integración de datos de fuentes diversas, aplicando técnicas avanzadas de análisis y modelado. Este enfoque tiene como objetivo optimizar la planificación urbana y promover sistemas de transporte más sostenibles en entornos urbanos, adaptándose a las necesidades cambiantes de las ciudades modernas.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Counting Time Temporal Logic for Multi-Robot Path Planning in Finite Horizons

  • Lv, Peng
  • Li, Shaoyuan
  • Mahulea, Cristian
  • Denis, Bruno
  • Faraut, Gregory
  • Yin, Xiang
In this paper, we consider multi-robot path planning problems for high-level tasks with a finite horizon. In many situations, there is a need to count how many times a sub-task is satisfied in order to achieve the overall task. However, existing temporal logic languages, such as linear temporal logic, is not efficient in describing such requirements. To address this issue, we propose a new temporal logic language called Counting Time Temporal Logic (CTTL) that extends linear temporal logic by explicitly counting the number of times that some tasks are satisfied. To solve the CTTL path planning problem, we propose an efficient integer linear programming-based method to encode task satisfaction. We show that our approach is both sound and complete, while achieving higher efficiency than direct encodings of such requirements. Moreover, we study several variants of the problem. To validate our results, we present several numerical experiments to show the scalability of the proposed approach and a simulation case study of a team of autonomous robots to illustrate the feasibility of the synthesis procedure. Finally, to evaluate the real-world feasibility of our method, we conduct a hardware experiment with two Turtlebot3-Burger mobile robots.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Recommending needles in a haystack: the SURGE approach

  • Hermoso, Ramon
  • Ilarri, Sergio
  • Trillo-Lado, Raquel
  • Marzo, Cristina
Collective Spatial Keyword Querying (CoSKQ) was proposed over a decade ago as a model to retrieve sets of objects in spatial databases given a specific query. The rationale behind is that the retrieved solution sets must cover query keywords as well as minimise the geographic distances between the query and the solution elements. However, in most real scenarios, the exact matching of query keywords and object descriptions is rare or not possible. In this paper, we extend the notion of CoSKQ for recommendation problems and present SURGE (Spatial User Recommendations using Geographical metrics and sEmantics), an approach that puts forward a recommender system able to return solution sets even when query keywords and object descriptions do not match exactly. In order to do that, semantic techniques are used. A tourism domain has been used throughout the paper to explain the model. Furthermore, an exhaustive set of experiments has been carried out to validate the approach.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

A Systematic Review of Cross-Population Shifts in Medical Imaging Analysis with Deep Learning

  • Musa, Aminu
  • Prasad, Rajesh
  • Onwualu, Peter
  • Hernández, Mónica
Deep learning has achieved expert-level performance in medical imaging analysis. However, models often fail to generalize across patient populations due to cross-population domain shifts, distributional differences arising from demographic variability, variations in imaging protocols, scanner hardware, and differences in disease prevalence. This challenge limits the real-world deployment and can increase health inequities. This review systematically examines the nature, causes, and impact of cross-population domain shift in deep learning-based medical imaging analysis. We analyzed 50 peer-reviewed studies from 2020 to 2025, evaluating the proposed methodologies for handling population shifts, the datasets employed, and the metrics used to assess performance. Our findings demonstrate that performance degradation ranged from 10–25% when models were tested on unseen populations, emphasizing the substantial impact of domain shifts on model generalizability. The literature reveals that mitigation strategies broadly fall into two categories: data-centric approaches, such as augmentation and harmonization, and model-centric approaches, including domain adaptation, transfer learning, adversarial learning, multi-task learning, and continual learning. While domain adaptation and transfer learning are the most widely used, their performance gains across populations remain modest, ranging from 5–15%, and are not supported by external validation. Our synthesis reveals a significant reliance on large, publicly available datasets from limited regions, with an underrepresentation of data from low- and middle-income countries. Evaluation practices are inconsistent, with few studies employing standardized external test sets. This review provides a structured taxonomy of mitigation techniques, a refined analysis of domain shift characteristics, and an in-depth critique of methodological challenges. We highlight the urgent need for more geographically and demographically inclusive datasets, adaptable modeling techniques, and standardized evaluation protocols to enable accurate and equitable AI-driven diagnostics across diverse populations. Finally, we outline future research directions to guide the development of robust, generalizable, and fair models for medical imaging analysis. El aprendizaje profundo ha alcanzado un rendimiento de nivel experto en el análisis de imágenes médicas. Sin embargo, los modelos a menudo no logran generalizar entre poblaciones de pacientes debido a cambios de dominio entre poblaciones, diferencias de distribución derivadas de la variabilidad demográfica, variaciones en los protocolos de imagen, el hardware del escáner y diferencias en la prevalencia de enfermedades. Este desafío limita la implementación en el mundo real y puede aumentar las inequidades en salud. Esta revisión examina sistemáticamente la naturaleza, las causas y el impacto de los cambios de dominio entre poblaciones en el análisis de imágenes médicas basado en aprendizaje profundo. Analizamos 50 estudios revisados ​​por pares de 2020 a 2025, evaluando las metodologías propuestas para manejar los cambios de población, los conjuntos de datos empleados y las métricas utilizadas para evaluar el rendimiento. Nuestros hallazgos demuestran que la degradación del rendimiento osciló entre el 10 % y el 25 % cuando los modelos se probaron en poblaciones no vistas, lo que enfatiza el impacto sustancial de los cambios de dominio en la generalización del modelo. La literatura revela que las estrategias de mitigación se dividen en dos categorías principales: enfoques centrados en datos, como la ampliación y la armonización, y enfoques centrados en modelos, que incluyen la adaptación de dominio, el aprendizaje por transferencia, el aprendizaje adversario, el aprendizaje multitarea y el aprendizaje continuo. Si bien la adaptación de dominio y el aprendizaje por transferencia son los más utilizados, sus mejoras de rendimiento en diferentes poblaciones siguen siendo modestas, oscilando entre el 5 % y el 15 %, y carecen de validación externa. Nuestra síntesis revela una dependencia significativa de grandes conjuntos de datos públicos procedentes de regiones limitadas, con una escasa representación de datos de países de ingresos bajos y medios. Las prácticas de evaluación son inconsistentes, y pocos estudios emplean conjuntos de prueba externos estandarizados. Esta revisión proporciona una taxonomía estructurada de las técnicas de mitigación, un análisis refinado de las características del cambio de dominio y una crítica exhaustiva de los desafíos metodológicos. Destacamos la necesidad urgente de contar con conjuntos de datos más inclusivos desde el punto de vista geográfico y demográfico, técnicas de modelado adaptables y protocolos de evaluación estandarizados para permitir diagnósticos precisos y equitativos basados ​​en IA en diversas poblaciones. Por último, describimos las futuras líneas de investigación para guiar el desarrollo de modelos robustos, generalizables y justos para el análisis de imágenes médicas.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Individual and collective gains from cooperation and reciprocity in a dynamic-network Prisoner’s Dilemma driven by extraversion, openness, and agreeableness

  • Abián, David
  • Bernad, Jorge
  • Ilarri, Sergio
  • Trillo-Lado, Raquel
How do stable personality differences shape cooperation when social ties can form and dissolve? We model a repeated Prisoner’s Dilemma on an endogenous network in which three continuous Big Five traits map to transparent local mechanisms: Extraversion sets a target number of partners, Openness determines how broadly agents search beyond friends-of-friends, and Agreeableness sets a baseline willingness to cooperate. At each encounter, agents combine this baseline with the partner’s directly observed history; there are no trait labels, gossip, or global reputations. Ties form when agents are under-connected and are cut when they become over-connected, with cuts prioritising partners who have defected more often. We vary network size (N=30–200), population composition, and the balance between trait-driven and history-driven behaviour. Three robust patterns emerge. First, cooperate first, then reciprocate—high initial willingness to cooperate combined with history-sensitive response—produces systems that are simultaneously more prosperous, fairer, and safer. Second, personality has predictable conditional effects: Agreeableness helps when history matters but hurts when behaviour is mostly trait-driven; Extraversion amplifies the environment; Openness has little net payoff effect. Third, the network reorganises accordingly: degree assortativity stays near zero, whereas agreeable agents increasingly connect to one another when cooperation takes hold.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publicación Artículo científico (JournalArticle)

Understanding Trade-Offs in Continuous Neural Representations for Diffeomorphic Image Registration: A Comparative Study of Implicit Neural Representations and Neural Ordinary Differential Equations

  • Rodriguez-Sanz, Salvador
  • Paesa-Lia, Carlos
  • Hernandez, Monica
Non-rigid image registration is a fundamental problem in medical imaging and a representative example of continuous transformation modeling in image processing. Diffeomorphic registration methods, such as Large Deformation Diffeomorphic Metric Mapping (LDDMM) and its PDE-constrained variants (PDE-LDDMM), provide mathematically grounded formulations with strong geometric guarantees for transformation quality. However, existing approaches face persistent trade-offs between numerical stability, accuracy, and computational efficiency. Recent work has explored implicit neural representations (INRs) and neural ordinary differential equations (NODEs) as flexible neural representations for modeling continuous transformations. Despite their increasing adoption, their practical behavior and limitations in diffeomorphic registration remain insufficiently understood. In this paper, we present a unified formulation of INR- and NODE-based registration methods within LDDMM and PDE-LDDMM, enabling a systematic and controlled comparison across architectures, sampling strategies, and numerical solvers. Our analysis reveals fundamental trade-offs between these approaches. In particular, we show that MLP-based INR formulations introduce significant computational overhead and rely on sampling strategies that can degrade smoothness and lead to the increased occurrence of non-diffeomorphic transformations at higher resolutions. Moreover, these approximations do not fully alleviate the computational cost, with some variants exceeding the costs of expensive classical optimization-based methods. In contrast, NODE-based formulations and downsampling strategies consistently provide transformations with more controlled Jacobian extrema while maintaining competitive computational performance. Among the evaluated methods, the original NODE-LDDMM and NODE-PDE-LDDMM formulations achieve the most favorable trade-offs between registration accuracy, geometric consistency, and computational efficiency. These findings provide clear insights into the design of neural representations for continuous transformation modeling, with practical implications for diffeomorphic registration and computational anatomy applications.