Pasar al contenido principal

INTELIGENCIA ARTIFICIAL EXPLICABLE PARA EL ENVEJECIMIENTO SALUDABLE

PID2021-123152OB-C21

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal para Impulsar la Investigación Científico-Técnica y su Transferencia
Subprograma Subprograma Estatal de Generación de Conocimiento
Convocatoria Proyectos de I+D+i (Generación de Conocimiento y Retos Investigación)
Año convocatoria 2021
Plan Estatal 2021-2023 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSIDADE DE SANTIAGO DE COMPOSTELA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 27
Encontrada(s) 1 página(s)

Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Artículo científico (JournalArticle)

Introducing User Feedback-based Counterfactual Explanations (UFCE)

  • Suffian, Muhammad
  • Alonso Moral, José María
  • Bogliolo, Alessandro
Machine learning models are widely used in real-world applications. However, their complexity makes it often challenging to interpret the rationale behind their decisions. Counterfactual explanations (CEs) have emerged as a viable solution for generating comprehensible explanations in eXplainable Artificial Intelligence (XAI). CE provides actionable information to users on how to achieve the desired outcome with minimal modifications to the input. However, current CE algorithms usually operate within the entire feature space when optimizing changes to turn over an undesired outcome, overlooking the identification of key contributors to the outcome and disregarding the practicality of the suggested changes. In this study, we introduce a novel methodology, that is named as user feedback-based counterfactual explanation (UFCE), which addresses these limitations and aims to bolster confidence in the provided explanations. UFCE allows for the inclusion of user constraints to determine the smallest modifications in the subset of actionable features while considering feature dependence, and evaluates the practicality of suggested changes using benchmark evaluation metrics. We conducted three experiments with five datasets, demonstrating that UFCE outperforms two well-known CE methods in terms of proximity, sparsity, and feasibility. Reported results indicate that user constraints influence the generation of feasible CEs.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Artículo científico (JournalArticle)

Physical activity phenotypes and mortality in older adults: a novel distributional data analysis of accelerometry in the NHANES

  • Matabuena Rodríguez, Marcos
  • Félix Lamas, Paulo
  • Hammouri, Ziad Akram Ali
  • Mota, Jorge
  • Pozo Cruz, Borja del
Physical activity is deemed critical to successful ageing. Despite evidence and progress, there is still a need to determine more precisely the direction, magnitude, intensity, and volume of physical activity that should be performed on a daily basis to effectively promote the health of individuals. This study aimed to assess the clinical validity of new physical activity phenotypes derived from a novel distributional functional analysis of accelerometer data in older adults. A random sample of participants aged between 65 and 80 years with valid accelerometer data from the National Health and Nutrition Examination Survey (NHANES) 2011–2014 was used. Five major clinical phenotypes were identified, which provided a greater sensitivity for predicting 5-year mortality and survival outcomes than age alone, and our results confirm the importance of moderate-to-vigorous physical activity. The new clinical physical activity phenotypes are a promising tool for improving patient prognosis and for directing to more targeted intervention planning, according to the principles of precision medicine. The use of distributional representations shows clear advantages over more traditional metrics to explore the effects of the full spectrum of the physical activity continuum on human health
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Artículo científico (JournalArticle)

AI-based emotion recognition in dementia through facial expression: A scoping review

  • Gerbaudo González, Noelia
  • Condori Fernández, Nelly
  • Catalá Bolós, Alejandro
  • Gandoy Crego, Manuel
Emotion assessment in dementia care is vital for patient well-being and effective care planning. Traditional methods are often subjective and time-consuming. This study examines the use of AI-based facial expression analysis for emotion recognition in dementia patients. A scoping review was conducted using the SPIDER strategy. Five databases—PubMed, Scopus, PsycInfo, ProQuest, and IEEE Xplore —were consulted, with additional records identified through snowballing. Data on participant characteristics, intervention details, non-AI comparisons, and clinical outcomes were categorized. Two authors independently screened records and extracted data on AI driven tools. The review analyzed 11 studies, primarily using deep neural networks. While most studies relied on pre-existing datasets, some collected original data. The studies focused on assessing a variety of emotions, with an emphasis on detecting basic emotions and, in some cases, more complex emotional states. AI applications included early detection, diagnosis, intervention impact assessment, and reliability testing. Comparisons were made with traditional assessment tools. This scoping review highlights the potential of AI tools to improve dementia care. However, standardized data collection and processing protocols are needed to advance AI in emotion recognition for dementia patients. Integrating multiple data sources and addressing dataset limitations are crucial for improving model accuracy and representativeness. Ethical considerations, including privacy and data security, must be prioritized when developing and implementing AI tools in this population. Interdisciplinary collaboration is essential to fully harness their potential
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Capítulo o parte de un libro (BookPart)

The role of Speculations for Explainable and Trustworthy Artificial Intelligence: A use case on Art Genre Classification

  • Alonso Moral, José María
  • Costa, Vicent
In this work, we have introduced a new way of speculative reasoning for intelligent systems. In addition, we have illustrated the utility of this way of reasoning in the context of a use case on art genre classification, where explainability and trustworthiness are a matter of major concern. Speculative reasoning is natural for humans and it turns up as a powerful tool, especially in the case of intelligent systems dealing with incomplete, vague, and imprecise information.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Capítulo o parte de un libro (BookPart)

Generating trustworthy explanations with a language model enriched by fuzzy rule-based systems

  • Pérez-Ferreiro, Pablo Miguel
  • Catalá Bolós, Alejandro
  • Bugarín-Diz, Alberto
  • Alonso Moral, José María
In this work, we have developed, validated, and deployed fuzzy-grounded, trustworthy, interactive explanations by combining fuzzy systems with Large Language Models (LLMs). More precisely, we profit from fuzzy systems' ability to deal appropriately with vagueness and uncertainty in approximate reasoning and propose an efficient and effective way to combine fuzzy rule-based systems with the natural language verbalization capabilities of LLMs. The aim is to generate explanations that balance naturalness and faithfulness. The proposed software architecture is open-source and highly modular. We validated our work in the medical field. We observed significant statistical differences with related-samples Wilcoxon signed rank tests (p-value < 0.001) regarding hallucinations in the experimental scenarios under study, which hints at good faithfulness. We also produced significantly more compact narratives, which helps alleviate the usual ‘over-explanation’ issue of LLMs without a penalization in accuracy.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Capítulo o parte de un libro (BookPart)

Evaluation of an Emotion-Aware Persuasive Framework Based on Peripheral Interaction for Reducing Physical Strain in Office Environments

  • Suni-Lopez, Franci
  • Condori Fernández, Nelly
Repetitive Strain Injuries (RSIs) are a leading cause of workplace disability and absenteeism, particularly among office workers engaged in prolonged computer-based tasks. This study evaluates the effectiveness of EMotivA, a persuasive system designed to promote active breaks and mitigate the risk of RSIs. A quantitative pretest-postest design was conducted with 20 office workers over one workweek, during which EMotivA dispatched persuasive notifications every 40 minutes to encourage brief, light physical activities. The system aimed to increase the frequency and regularity of active breaks and reduce musculoskeletal discomfort, measured using the Cornell Musculoskeletal Discomfort Questionnaire (CMDQ). Results showed a significant increase in active breaks and a marked improvement in time dedicated to physical activities. CMDQ scores revealed a significant reduction in musculoskeletal discomfort post-intervention (t(19) = 3.152, p = 0.005, d = 0.705). These findings demonstrate that EMotivA effectively fosters healthier work habits and reduces physical discomfort, emphasizing its potential as a tool for occupational health management.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Artículo científico (JournalArticle)

Enriching interactive explanations with fuzzy temporal constraint networks

  • Canabal Juanatey, Mariña
  • Alonso Moral, José María
  • Catalá Bolós, Alejandro
  • Bugarín-Diz, Alberto
Humans often use expressions with vague terms which play a fundamental role for effective communication. These expressions are successfully modeled with fuzzy technology, but they are not usually integrated yet with Natural Language Processing models and techniques. Large-scale pre-trained language models yield excellent results in many language tasks, but they have some drawbacks such as their lack of transparency and thorough temporal reasoning capabilities. Therefore, the use of such models may provoke inconsistent or incorrect dialogues in the context of conversational agents which were aimed at providing users of intelligent systems with interactive explanations. In this paper, we propose a model for fuzzy temporal reasoning to overcome some inconsistencies detected in pre-trained language models in a specific application domain of a conversational agent carefully designed for providing users with explanations which are endowed with a good balance between naturalness and fidelity. More precisely, starting from a knowledge graph that provides an intuitive representation of the entities and relations in the application domain, we describe how to map the temporal information onto a fuzzy temporal constraint network. This formalism allows to represent imprecise temporal information and provides mechanisms for checking consistency in conversations. In addition, as a proof of concept, we have developed TimeVersa, a conversational agent which integrates the proposed model into an application domain (i.e., a virtual assistant for tourists) that requires handling imprecise temporal constraints. We illustrate in a use case how the agent can identify temporal inconsistencies and answer queries related to temporal information properly. Results after a user study report that users' perception of consistency is significantly higher in a conversation with TimeVersa than in a similar conversation using the well-known GPT-3 Large Language Model, when vague temporal information is involved. The proposed approach is a step forward for developing conversational agents operating in application domains that require temporal reasoning under uncertainty.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Capítulo o parte de un libro (BookPart)

A User Study on the Utility of Context-aware Explanations for Assisting Data Scientists in Error Analysis of Fuzzy Decision Trees

  • Fernández, Guillermo
  • Gámez, Jose A.
  • Puerta, Jose M.
  • Aledo, Juan A.
  • Alonso Moral, José María
  • Bugarín-Diz, Alberto
The current growth of interest in the field of eXplainable Artificial Intelligence has led to the rise of multiple explanation generation techniques to bridge the gap between the algorithmic complexity of the most powerful algorithms and their end users, who are expected to take advantage of them. In addition, explanations can assist with other tasks, such as explaining the algorithm to a model designer, who can use that information to fine-tune the system that implements such an algorithm. This gives the model designer some insight into the inner workings of the system to guide design decisions as well as identify and prevent potential errors, resulting in a myriad of improvements regarding model properties such as accuracy, explainability, trustworthiness, coherence with existing knowledge, etc. In this paper, we introduce a method to enrich local explanations of fuzzy decision tree classifications with context information. The goodness of the proposed method was validated with a user study in which 26 participants had to detect classification errors made by a fuzzy decision tree. Reported results show that enriched explanations allowed the participants to be more accurate in detecting errors in the classification system, albeit taking longer to solve the task with respect to participants who were shown regular local explanations.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Capítulo o parte de un libro (BookPart)

CL-XAI: Toward Cognitive Learning through eXplainable Artificial Intelligence

  • Suffian, Muhammad
  • Kuhl, Ulrike
  • Alonso Moral, José María
  • Bogliolo, Alessandro
Artificial Intelligence (AI) is transforming education by providing personalized learning paths that cater to individual needs. Explainable AI (XAI) plays a vital role in assisting learners with limited prior knowledge and skills, offering transparency by elucidating how AI systems reach conclusions. In a co-learning environment, where learners engage collaboratively, the role of XAI becomes even more significant. This transparency not only fosters trust in AI-driven learning but also empowers co-learners to actively participate in and enhance their cognitive learning journey. Providing explanations for novel concepts is recognised as a fundamental aid in the learning process, particularly when addressing challenges stemming from knowledge deficiencies and skill application. Addressing these difficulties involves timely explanations and guidance throughout the learning process, prompting the interest of AI experts in developing explainer models. In this paper, we introduce an intelligent system (CL-XAI) for cognitive learning supported by XAI, focusing on two key research objectives: (i) exploring how human learners comprehend the internal mechanisms of AI models using XAI tools; and (ii) evaluating the effectiveness of such tools through human feedback. The use of CL-XAI is illustrated with a game-inspired virtual use case where learners tackle combinatorial problems to enhance problem-solving skills and deepened their understanding of complex concepts. The analysis of a pilot study of 21 participants indicates improved learning outcomes with explanations, showing a positive correlation with the complexity of tasks. Notably, over 60% of the participants expressed satisfaction, recognizing explanations as valuable aids in achieving learning targets. This paper is a pathway highlighting the potential for transformative advances in cognitive learning and co-learning.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Capítulo o parte de un libro (BookPart)

Beyond Prediction Similarity: ShapGAP for Evaluating Faithful Surrogate Models in XAI

  • Mariotti, Ettore
  • Sivaprasad, Adarsa
  • Alonso Moral, José María
The growing importance of Explainable Artificial Intelligence (XAI) has highlighted the need to understand the decision-making processes of black-box models. Surrogation, emulating a black-box model (BB) with a white-box model (WB), is crucial in applications where BBs are unavailable due to security or practical concerns. Traditional fidelity measures only evaluate the similarity of the final predictions, which can lead to a significant limitation: considering a WB faithful even when it has the same prediction as the BB but with a completely different rationale. Addressing this limitation is crucial to develop Trustworthy AI practical applications beyond XAI. To address this issue, we introduce ShapGAP, a novel metric that assesses the faithfulness of surrogate models by comparing their reasoning paths, using SHAP explanations as a proxy. We validate the effectiveness of ShapGAP by applying it to real-world datasets from healthcare and finance domains, comparing its performance against traditional fidelity measures. Our results show that ShapGAP enables better understanding and trust in XAI systems, revealing the potential dangers of relying on models with high task accuracy but unfaithful explanations. ShapGAP serves as a valuable tool for identifying faithful surrogate models, paving the way for more reliable and Trustworthy AI applications.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Artículo científico (JournalArticle)

An Empirical Study on the Number of Items in Human Evaluation of Automatically Generated Texts

  • González-Corbelle, Javier
  • Alonso Moral, José María
  • Crujeiras Casais, Rosa María
  • Bugarín-Diz, Alberto
Human evaluation of neural models in the research field of Natural Language Generation (NLG) requires a careful experimental design (e.g., number of evaluators, number of items to assess, number of quality criteria, etc.) for the sake of reproducibility as well as for ensuring that it is possible to draw significant conclusions. Although there are some generic recommendations on how to proceed, there is not an evaluation protocol admitted worldwide. In this paper, we address empirically the impact of the number of items to assess in the context of the human evaluation of NLG systems. We apply different resampling methods to simulate the evaluation of different sets of items by each evaluator. Then, we compare the results obtained by evaluating only a limited set of items with those obtained by evaluating all outputs of the system for a given test set. Empirical findings validate the initial research hypothesis: well-known resampling statistical methods can contribute to getting significant results even with a small number of items to be evaluated by each evaluator.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Artículo científico (JournalArticle)

Trustworthy Artificial Intelligence in Alzheimer’s Disease: State of the Art, Opportunities, and Challenges

  • El-Sappagh, Shaker
  • Alonso Moral, José María
  • Abuhmed, Tamer
  • Ali, Farman
  • Bugarín-Diz, Alberto
Medical applications of Artificial Intelligence (AI) have consistently shown remarkable performance in providing medical professionals and patients with support for complex tasks. Nevertheless, the use of these applications in sensitive clinical domains where high-stakes decisions are involved could be much more extensive if patients, medical professionals, and regulators were provided with mechanisms for trusting the results provided by AI systems. A key issue for achieving this is endowing AI systems with key dimensions of Trustworthy AI (TAI), such as fairness, transparency, robustness, or accountability, which are not usually considered within this context in a generalized and systematic manner. This paper reviews the recent advances in the TAI domain, including TAI standards and guidelines. We propose several requirements to be addressed in the design, development, and deployment of TAI systems and present a novel machine learning pipeline that contains TAI requirements as embedded components. Moreover, as an example of how current AI systems in medicine consider the TAI perspective, the study extensively reviews the recent literature (2017-2021) on AI systems in a prevalent and high social-impact disease: diagnosis and progression detection of Alzheimer’s Disease (AD). The most relevant AI systems in the AD domain are compared and discussed (such as machine learning, deep learning, ensembles, time series, and multimodal multitask) from the perspective of how they address TAI in their design. Several open challenges are highlighted, which could be claimed as one of the main reasons to justify the rare application of AI systems in real clinical environments. The study provides a roadmap to measure the TAI status of an AI systems and highlights its limitations. In addition, it provides the main guidelines to overcome these limitations and build medically trusted AI-based applications in the medical domain.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Capítulo o parte de un libro (BookPart)

Logs Based Verification Tool of Serious Game for Autistic Children

  • Nur Rohmah, Arini
  • Condori Fernández, Nelly
Serious games are referred to as entertaining tools with a purpose of education. The correct implementation of serious games is very important to ensure a high level of adoption of the competencies by the game users, which can be even more challenging if we target autistic children. Understanding the actual needs of Children with Autism Spectrum Disorder (ASD) and supporting their learning activity through serious games can be a burden. This paper presents the architecture of logs based Verification tool, named GVT, to identify potential scenes of the serious game that might be causing any negative emotion on the autistic children. GVT architecture follows a client sever model, which enables to merge the game logs with the emotional data logs. The data analysis and visualization, carried out at the server side, should provide relevant information to support developers to enhance the quality of serious games for autistic people.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Artículo científico (JournalArticle)

An empirical study on how humans appreciate automated counterfactual explanations which embrace imprecise information

  • Stepin, Ilia
  • Pereira Fariña, Martín
  • Alonso Moral, José María
  • Catalá Bolós, Alejandro
The explanatory capacity of interpretable fuzzy rule-based classifiers is usually limited to offering explanations for the predicted class only. A lack of potentially useful explanations for non-predicted alternatives can be overcome by designing methods for the so-called counterfactual reasoning. Nevertheless, state-of-the-art methods for counterfactual explanation generation require special attention to human evaluation aspects, as the final decision upon the classification under consideration is left for the end user. In this paper, we first introduce novel methods for qualitative and quantitative counterfactual explanation generation. Then, we carry out a comparative analysis of qualitative explanation generation methods operating on (combinations of) linguistic terms as well as a quantitative method suggesting precise changes in feature values. Then, we propose a new metric for assessing the perceived complexity of the generated explanations. Further, we design and carry out two human evaluation experiments to assess the explanatory power of the aforementioned methods. As a major result, we show that the estimated explanation complexity correlates well with the informativeness, relevance, and readability of explanations perceived by the targeted study participants. This fact opens the door to using the new automatic complexity metric for guiding multi-objective evolutionary explainable fuzzy modeling in the near future
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Artículo científico (JournalArticle)

Trustworthy Data-driven Chronological Age Estimation from Panoramic Dental Images

  • Vivel Couso, Ainhoa
  • Vila Blanco, Nicolás
  • Carreira Nouche, María José
  • Bugarín-Diz, Alberto
  • Tomás Carmona, Inmaculada
  • Alonso Moral, José María
Integrating deep learning into healthcare enables personalized care but raises trust issues due to model opacity. To improve transparency, we propose a system for dental age estimation from panoramic images that combines an opaque and a transparent method within a natural language generation (NLG) module. This module produces clinician-friendly textual explanations about the age estimations, designed with dental experts through a rule-based approach. Following the best practices in the field, the quality of the generated explanations was manually validated by dental experts using a questionnaire. The results showed a strong performance, since the experts rated 4.77$\pm$0.12 (out of 5) on average across the five dimensions considered. We also performed a trustworthy self-assessment procedure following the ALTAI checklist, in which it scored 4.40$\pm$0.27 (out of 5) across seven dimensions of the AI Trustworthiness Assessment List.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Capítulo o parte de un libro (BookPart)

AffeX: a real-time emotion monitoring and visualization tool

  • Méndez Flórez, Juan Camilo
  • Condori Fernández, Nelly
  • Catalá Bolós, Alejandro
  • Alonso Moral, José María
  • Figueroa Vásquez, Salvador Alonso
Non-invasive, real-time emotion detection enables continuous and discreet monitoring of affective states, supporting system adaptability while safeguarding user privacy, an essential feature in clinical, educational, and research contexts. This paper presents AffeX, an intuitive tool for the real-time visualization and analysis of facial emotions, designed with a focus on privacy and data transparency. It offers customizable visual feedback, allowing the exploration of emotional responses without compromising identity. Unlike commercial solutions, AffeX ensures full control over data handling, enabling further analysis and multimodal integration, making it especially suitable for privacy-aware and adaptable research scenarios.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Artículo científico (JournalArticle)

TextFocus: Assessing the Faithfulness of Feature Attribution Methods Explanations in Natural Language Processing

  • Mariotti, Ettore
  • Arias-Duart, Anna
  • Cafagna, Michele
  • Gatt, Albert
  • García-Gasulla, Darío
  • Alonso Moral, José María
Among the existing eXplainable AI (XAI) approaches, Feature Attribution methods are a popular option due to their interpretable nature. However, each method leads to a different solution, thus introducing uncertainty regarding their reliability and coherence with respect to the underlying model. This work introduces TextFocus, a metric for evaluating the faithfulness of Feature Attribution methods for Natural Language Processing (NLP) tasks involving classification. To address the absence of ground truth explanations for such methods, we introduce the concept of textual mosaics. A mosaic is composed of a combination of sentences belonging to different classes, which provides an implicit ground truth for attribution. The accuracy of explanations can be then evaluated by comparing feature attribution scores with the known class labels in the mosaic. The performance of six feature attribution methods is systematically compared on three sentence classification tasks by using TextFocus, with Integrated Gradients being the best overall method in terms of faithfulness and computational requirements. The proposed methodology fills a gap in NLP evaluation, by providing an objective way to assess Feature Attribution methods while finding their optimal parameters.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Artículo científico (JournalArticle)

The Role of User Feedback in Enhancing Understanding and Trust in Counterfactual Explanations for Explainable AI

  • Suffian, Muhammad
  • Kuhl, Ulrike
  • Bogliolo, Alessandro
  • Alonso Moral, José María
Counterfactual explanations (CEs) have emerged as a viable solution for generating comprehensible explanations in the context of explainable artificial intelligence (XAI). A CE provides actionable information to users on how to achieve the desired outcome from a machine learning (ML) model with minimal modifications to the input. XAI is crucial for improving transparency and reliability in AI systems, especially for meeting regulations like the General Data Protection Regulation (GDPR) or the European AI Act. However, the integration of CEs into XAI frameworks and their effectiveness in enhancing user trust and cognitive learning remains uncertain and requires further research. We have developed a user study to face this challenge with two user input-driven counterfactual generation XAI approaches: (i) User Feedback-based Counterfactual Explanation (UFCE) and (ii) Diverse Counterfactual Explanation (DiCE). They are integrated within a game-inspired online platform that enables direct comparisons between them.We compared the task performance, understanding, satisfaction, and trust between control and experimental groups, with a total of 101 participants. After curating the collected data, we had 70 users (24 in the control group) who successfully completed the experiment. Participants in the experimental group received explanations generated by UFCE or DiCE. Findings show that explanations generated by UFCE improve users’ learning experiences, resulting in better task performance, comprehension, satisfaction, and trust. Moreover, participants who interacted with UFCE exhibited significantly higher reliance on suggestions than those who interacted with DiCE, what was supported by statistical validation. These results highlight the significance of human-centered XAI methods and promote meaningful cognitive engagement for users. Furthermore, the game-inspired platform is implemented as open-source to promote Open Science, and it is made publicly available along with data collected in the user study to support further investigations and to ensure reproducibility of reported results.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Artículo científico (JournalArticle)

Adapting Technology for Dementia Care: The Case of Emobook App in Reminiscence focused Music Therapy

  • Gerbaudo González, Noelia
  • Catalá Bolós, Alejandro
  • Condori Fernández, Nelly
  • Gandoy Crego, Manuel
INTRODUCTION: Life Story books are frequently employed to facilitate reminiscence interventions, but their use in Music Therapy remains limited in the scientific literature. There is a paucity of research detailing the design processes involved in this context. OBJECTIVES: This paper aims to report on the adaptation of the Emobook Life Story Book App for a Reminiscence Music Therapy Program for people living with dementia. METHODS: An interdisciplinary team comprising an interaction designer, a software engineer, a music therapist, and a research assistant engaged in a Participatory Design process. The study comprised two distinct phases: "Adaptation Phase," aimed to identify features requiring adjustment to integrate Emobook into MT effectively and the "Implementation Phase," focused on evaluating the adequacy of these changes for Emobook's use within the Music Therapy Program RESULTS: By merging the requirements identified during each phase, additional improvements were generated, leading to the decision to evolve Emobook Post-Prototype towards a version tailored specifically for Music Therapy. CONCLUSION: Collaborative, interdisciplinary efforts are essential in advancing the incorporation of technology into music therapy practice. This study demonstrates the value of a Participatory Design approach in this regard
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Artículo científico (JournalArticle)

Exploring the balance between interpretability and performance with carefully designed constrainable Neural Additive Models

  • Mariotti, Ettore
  • Alonso Moral, José María
  • Gatt, Albert
The interpretability of an intelligent model automatically derived from data is a property that can be acted upon with a set of structural constraints that such a model should adhere to. Often these are in contrast with the task objective and it is not straightforward how to explore the balance between model interpretability and performance. In order to allow an interested user to jointly optimise performance and interpretability, we propose a new formulation of Neural Additive Models (NAM) which can be subject to a number of constraints. Accordingly, our approach produces a new model that is called Constrainable NAM (or just CNAM in short) and it allows the specification of different regularisation terms. CNAM is differentiable and is built in such a way that it can be initialised as a solution of an efficient tree-based GAM solver (e.g., Explainable Boosting Machines). From this local optimum the model can then explore solutions with different interpretability-performance tradeoffs according to different definitions of both interpretability and performance. We empirically benchmark the model on 56 datasets against 12 models and observe that on average the proposed CNAM model ranks on the Pareto front of optimal solutions, i.e., models generated by CNAM exhibit a good balance between interpretability and performance. Moreover, we provide two illustrative examples which are aimed to show step by step how CNAM works well for solving classification tasks, but also how it can yield insights when considering regression tasks
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Artículo científico (JournalArticle)

AI literacy in K‑12: a systematic literature review

  • Casal Otero, Lorena
  • Catalá Bolós, Alejandro
  • Fernández Morante, Carmen
  • Taboada Iglesias, María Jesús
  • Cebreiro López, Beatriz
  • Barro Ameneiro, Senén
The successful irruption of AI-based technology in our daily lives has led to a growing educational, social, and political interest in training citizens in AI. Education systems now need to train students at the K-12 level to live in a society where they must interact with AI. Thus, AI literacy is a pedagogical and cognitive challenge at the K-12 level. This study aimed to understand how AI is being integrated into K-12 education worldwide. We conducted a search process following the systematic literature review method using Scopus. 179 documents were reviewed, and two broad groups of AI literacy approaches were identified, namely learning experience and theoretical perspective. The first group covered experiences in learning technical, conceptual and applied skills in a particular domain of interest. The second group revealed that significant efforts are being made to design models that frame AI literacy proposals. There were hardly any experiences that assessed whether students understood AI concepts after the learning experience. Little attention has been paid to the undesirable consequences of an indiscriminate and insufficiently thought-out application of AI. A competency framework is required to guide the didactic proposals designed by educational institutions and define a curriculum reflecting the sequence and academic continuity, which should be modular, personalized and adjusted to the conditions of the schools. Finally, AI literacy can be leveraged to enhance the learning of disciplinary core subjects by integrating AI into the teaching process of those subjects, provided the curriculum is co-designed with teachers.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Capítulo o parte de un libro (BookPart)

Virtual Assistant supported by Fusion of Fuzzy Rule-based Explainers

  • Pérez-Ferreiro, Pablo Miguel
  • Catalá Bolós, Alejandro
  • Bugarín-Diz, Alberto
  • Alonso Moral, José María
In this work, we introduce a novel methodology for combining different explanatory inputs into a singular coherent narrative in the medical domain. As proof of concept, we have implemented a prototype ready to diagnose diabetes and heart disease. The proposal is validated with a user study in which we collected 48 valid responses to a given questionnaire. We compared four different strategies for explanation fusion, considering explanations coming from two different fuzzy rule bases. The winner strategy (i.e., verbalizing common antecedents as the main factors of the diagnosis and adding any non-common antecedents as supporting evidence) was perceived as trustworthy, useful, and natural by most participants in the study.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Capítulo o parte de un libro (BookPart)

A Confusion Matrix for Evaluating Feature Attribution Methods

  • Arias-Duart, Anna
  • Mariotti, Ettore
  • García-Gasulla, Darío
  • Alonso Moral, José María
The increasing use of deep learning models in critical areas of computer vision and the consequent need for insights into model behaviour have led to the development of numerous feature attribution methods. However, these attributions must be both meaningful and plausible to end-users, which is not always the case. Recent research has emphasized the importance of faithfulness in attributions, as plausibility without faithfulness can result in misleading explanations and incorrect decisions. In this work, we propose a novel approach to evaluate the faithfulness of feature attribution methods by constructing an ‘Attribution Confusion Matrix’, which allows us to leverage a wide range of existing metrics from the traditional confusion matrix. This approach effectively introduces multiple evaluation measures for faithfulness in feature attribution methods in a unified and consistent framework. We demonstrate the effectiveness of our approach on various datasets, attribution methods, and models, emphasizing the importance of faithfulness in generating plausible and reliable explanations while also illustrating the distinct behaviour of different feature attribution methods.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Artículo científico (JournalArticle)

Measuring and implementing lexical alignment: A systematic literature review

  • Srivastava, Sumit
  • Wentzel, Suzanna D.
  • Catalá Bolós, Alejandro
  • Theune, Mariët
Lexical Alignment is a phenomenon often found in human–human conversations, where the interlocutors converge during a conversation to use the same terms and phrases for the same underlying concepts. Alignment (linguistic) is a mechanism used by humans for better communication between interlocutors at various levels of linguistic knowledge and features, and one of them is lexical. The existing literature suggests that alignment has a significant role in communication between humans, and is also beneficial in human–agent communication. Various methods have been proposed in the past to measure lexical alignment in human–human conversations, and also to implement them in conversational agents. In this research, we carry out an analysis of the existing methods to measure lexical alignment and also dissect methods to implement it in a conversational agent for personalizing human–agent interactions. We propose a new set of criteria that such methods should meet and discuss the possible improvements that can be made to existing methods.
Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Publicación Artículo científico (JournalArticle)

How to Build Self-Explaining Fuzzy Systems: From Interpretability to Explainability

  • Stepin, Ilia
  • Suffian, Muhammad
  • Catalá Bolós, Alejandro
  • Alonso Moral, José María
Fuzzy systems are known to provide not only accurate but also interpretable predictions. However, their explainability may be undermined if non-semantically grounded linguistic terms are used. Additional non-trivial challenges would arise if a prediction were to be explained counterfactually, i.e., in terms of hypothetical, non-predicted outputs. In this paper, we explore how both factual and counterfactual automated explanations can justify the output of fuzzy rule-based classifiers, and thus contribute to making them more trustworthy. Moreover, we demonstrate how end user preferences can be handled by customizing automated explanations, making them interactive, personalized, and therefore, human-centric. The full immersive article at IEEE Xplore provides detailed interactive examples for better understanding.
RODIN. Repositorio de Objetos de Docencia e Investigación de la Universidad de Cádiz
Publicación Artículo científico (JournalArticle)

Physical activity phenotypes and mortality in older adults: a novel distributional data analysis of accelerometry in the NHANES

  • Matabuena, Marcos
  • Félix, Paulo
  • Ali Hammour, Ziad Akram
  • Mota, Jorge
  • Pozo Cruz, Borja del
Physical activity is deemed critical to successful ageing. Despite evidence and progress, there is still a need to determine more precisely the direction, magnitude, intensity, and volume of physical activity that should be performed on a daily basis to effectively promote the health of individuals. This study aimed to assess the clinical validity of new physical activity phenotypes derived from a novel distributional functional analysis of accelerometer data in older adults. A random sample of participants aged between 65 and 80 years with valid accelerometer data from the National Health and Nutrition Examination Survey (NHANES) 2011–2014 was used. Five major clinical phenotypes were identified, which provided a greater sensitivity for predicting 5-year mortality and survival outcomes than age alone, and our results confirm the importance of moderate-to-vigorous physical activity. The new clini- cal physical activity phenotypes are a promising tool for improving patient prognosis and for directing to more targeted intervention planning, according to the principles of precision medicine. The use of distributional representations shows clear advantages over more traditional metrics to explore the effects of the full spectrum of the physical activity continuum on human health.
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

An art painting style explainable classifier grounded on logical and commonsense reasoning

  • Costa, Vicent
  • Alonso-Moral, José M.
  • Falomir, Zoe
  • Dellunde, Pilar
[Data Availability] The QArt-337 dataset, images, scripts needed for reproducibility of experiments as well as logs with reported results are available at https://gitlab.citius.usc.es/jose.alonso/xai4art/., This paper presents the art painting style explainable classifier named ANYXI. The classifier is based on art specialists’ knowledge of art styles and human-understandable color traits. ANYXI overcomes the principal flaws in the few art painting style classifiers in the literature. In this way, we first propose, using the art specialists’ studies, categorizations of the Baroque, Impressionism, and Post-Impressionism. Second, we carry out a human survey with the aim of validating the appropriateness of the color features used in the categorizations for human understanding. Then, we analyze and discuss the accuracy and interpretability of the ANYXI classifier. The study ends with an evaluation of the rationality of explanations automatically generated by ANYXI. We enrich the discussion and empirical validation of ANYXI by considering a quantitative and qualitative comparison versus other explainable classifiers. The reported results show how ANYXI is outstanding from the point of view of interpretability while keeping high accuracy (comparable to non-explainable classifiers). Moreover, automated generations are endowed with a good level of rationality., Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature. This work is supported by MCIN/AEI/10.13039/501100011033 under grants PID2021-123152OB-C21, PID2021-123152OB-C22, and RED2022-134315-T and the Galician Ministry of Culture, Education, Professional Training and University under grants ED431G2019/04 and ED431C2022/19. All grants were cofunded by the European Union and FEDER/ERDF (European Regional Development Funds). This work is also partially funded by the H2020-MSCARISE- 2020 project MOSAIC (Grant Agreement 101007627). Z. Falomir acknowledges funding by the Ramon-y-Cajal fellowship (RYC2019-027177-I / AEI / 10.13039/501100011033) awarded by the Spanish Ministry of Science, Innovation and Universities. V. Costa received funding from the Juan de la Cierva fellowship (FJC2021-047274-I) awarded by the Spanish Ministry of Science, Innovation and Universities, and the European Union (NextGenerationEU), and acknowledges the GEHUCT research group (2021 SGR 00517)., Peer reviewed