Jose Antonio Gallud

Digital platforms generate interaction logs that support behaviour-based user modelling when direct assessment is unavailable. However, when proxy labels and predictive features are derived from the same behavioural source, measured performance may depend on how the target is constructed, represented, and encoded in the predictors. This study examines these dependencies using HEXAD-inspired behavioural proxies derived from 1,037 Moodle users and 109,729 mapped interaction events. Two proxy-label constructions based on raw event frequencies and term frequency–inverse document frequency weighting were evaluated with corresponding feature representations, five supervised learning algorithms, optional in-fold Synthetic Minority Over-sampling Technique, and repeated stratified five-fold cross-validation with ten repeats. The proxy constructions assigned different labels to 136 users (13.11%) and produced different class distributions. Across matched configurations, changing only the proxy construction yielded observed macro-averaged F1 score differences of up to 0.0786, but no comparison remained statistically significant after Holm correction. The highest observed performance was obtained by Extreme Gradient Boosting using term frequency–inverse document frequency features, raw proxy labels, and in-fold oversampling, reaching a mean macro-averaged F1 score of 0.9147 and a mean balanced accuracy of 0.9610. Corrected comparisons did not establish an advantage over Random Forest. Deterministic analysis showed perfect recovery of term frequency–inverse document frequency proxy labels and perfect recovery of raw proxy labels for users with a unique raw maximum. These findings show that proxy-label construction defines the supervised task and that high predictive performance should be interpreted as recovery of engineered behavioural proxies rather than independent psychometric identification of latent motivational profiles.

Learner profiling is a key component of adaptive and personalized learning environments. While the HEXAD framework provides a theoretically grounded taxonomy of motivational user types, its practical use in learning management systems (LMS) typically relies on self‐reported questionnaires, which limits scalability and ecological validity. This study investigates whether meaningful learner profiles inspired by HEXAD can be derived directly from behavioral interaction data in a fully unsupervised manner. Using Moodle log data collected across multiple courses, learner activity is aggregated at the student level and represented through two alternative feature encodings: raw event frequencies and TF–IDF weighting. Hierarchical clustering is applied under consistent conditions, and clustering solutions are evaluated in terms of stability under subsampling, internal diagnostics for cluster resolution, and sensitivity to methodological choices. A cluster configuration with is examined as a theory‐informed reference aligned with the HEXAD framework. Results show that TF–IDF representations produce more stable and robust clustering structures than raw frequencies, particularly at the HEXAD‐aligned resolution. To support interpretability, a dictionary‐based surrogate profiling approach is used to characterise behavioral patterns associated with each cluster, without treating these surrogate labels as ground truth. The analysis reveals imbalances in profile prevalence and highlights the importance of representation choices in unsupervised learner modeling. These findings show that interpretable learner profiles can be derived retrospectively from naturally occurring LMS data without relying on questionnaires. They provide a methodological basis for behavior‐based learner modeling, while the feasibility of generating sufficiently stable profiles early enough for time‐sensitive adaptation remains to be evaluated.

Los agentes autónomos de IA están transformando la relación de las personas con sus datos, actuando como intermediarios que perciben, razonan y operan sobre la información con supervisión humana limitada. Este artículo investiga la convergencia de los principios de Interacción Humano-Datos (HDI) y las arquitecturas modernas de agentes IA, tomando como referencia el Agent Development Kit (ADK) de Google. A partir de los fundamentos de HDI —legibilidad, agencia y negociabilidad— introducimos un modelo triádico Humano-Agente-Datos (HAD) que captura los desafíos de las experiencias de datos mediadas por agentes. El análisis expone tensiones entre autonomía y control, transparencia y eficiencia de delegación, y gobernanza individual frente a colectiva. Traducimos estas ideas en siete principios de diseño centrados en el usuario y un conjunto de extensiones arquitectónicas para ADK (delegación graduada, responsabilidad epistémica y permisos negociables), y discutimos sus implicaciones para el RGPD, el AI Act y la ENIA 2.0. Se incluye un escenario ilustrativo, una comparativa de plataformas y un protocolo de evaluación piloto con usuarios que operacionaliza las métricas de legibilidad y preservación del control.

AI-assisted competency assessments increasingly depend on evolving technical components such as sensors, automated scoring mechanisms, machine-learning models, generative AI systems, and multimodal processing pipelines. Changes to these components create a lifecycle problem: technical compatibility does not ensure that previous evidential support remains applicable, while technical change does not necessarily require renewed validation of the whole assessment.This paper introduces a lifecycle framework for reconfigurable AI-assisted competency assessment combining two complementary artefacts. The \emph{Measurement--Evidence Bridge} (MEB) represents dependencies between technical realizations, measurement properties, assessment evidence, evidential warrants, contextual conditions, Validation Evidence, and Assessment Claims. The \emph{Evidential Change Impact Analysis} (ECIA) uses these dependencies to determine whether support established for one assessment configuration can be transferred after a change and where revalidation may be required. The framework thus distinguishes technical change, transfer of evidential support, and consequences for assessment claims.The framework is examined through a controlled analytical evaluation covering five types of change: sensor replacement, AI evidence-processor replacement, evidence-source removal, task change, and intended-use change. Results show that MEB and ECIA identify warrant-relevant impacts while preserving unaffected evidential relationships. They also show that task or intended-use changes may affect previous support even when the technical configuration remains unchanged. The framework therefore supports targeted, traceable revalidation requirements, avoiding assumptions that previous support automatically transfers or that every change requires complete revalidation. The evaluation supports the internal coherence and analytical feasibility of the approach, while further empirical evaluation in operational assessment settings is needed.

La motivación lectora en primaria es baja y las bibliotecas escolares están infrautilizadas, especialmente cuando no están integradas en los entornos virtuales de aprendizaje (VLE) y además se sufre una conectividad irregular, como ocurre en entornos rurales. Este artículo presenta LibroTech, una biblioteca escolar virtual gamificada integrada en MundoCRA que vincula artefactos propios de las bibliotecas, con reglas de juego para alinear objetivos de lectura con progreso y reconocimiento. Para ello se presenta el diseño de una arquitectura cliente–servidor con procesamiento asíncrono que permite gestionar analíticas y recompensas, flujos diferenciados por rol e integración visual optimizada. El trabajo incluye una evaluación exploratoria de usabilidad en aula (n = 9; con 6 estudiantes y 3 docentes) midiendo eficacia, eficiencia y satisfacción, junto a observación de latencia. Los resultados muestran una alta finalización de tareas, y una satisfacción de 84.5 en estudiantes y 86.67 en docentes. Del cuestionario se puede inferir también que la aprendibilidad es del 91.67%. En cuanto al impacto, esta investigación evidencia que la aceptabilidad y viabilidad en contextos rurales es satisfactoria y ha servido para delinear mejoras prioritarias (analítica ampliada, personalización y accesibilidad), con potencial para impulsar hábitos lectores entre los estudiantes.

This chapter presents the design, implementation, and evaluation of TEA Minigames, a customizable system of serious games to promote social inclusion for children with Autism Spectrum Disorder (ASD). It includes a mobile app with three minigames –emotion recognition, daily routine sequencing, and social reasoning– plus a web interface for caregivers and educators to adapt content. Developed with a local autism association, it follows accessibility, universal design, and user-centered principles, addressing cognitive and sensory needs through flexible content, minimal interface, immediate feedback, and an evaluation protocol based on ISO/IEC usability standards. Pilot testing with caregivers and children confirmed the system is engaging, usable, and appropriate for educational and therapeutic use. The chapter closes with implications for accessible educational software design and future directions for expansion and inclusive interaction.

In spite of all advances promoted by information technologies, there are still activities where this technology is not applied for reasons such as being carried out in non-profit organizations or because they have not adapted to this modernization. Until recently, the way to work with mobile devices was either by connecting through a web page with the device's browser, or by downloading an application from the corresponding platform. But lately, technologies are being developed that aim to break with this, as in the case of Progressive Web Applications (PWA). One of the advantages offered by PWA is to access the web page and install it as an application on the device. The purpose of this article is to design a progressive Web application for the support of Storytelling Therapy, one of the novel therapies applied in the field of mental health. In addition to providing a software application to enhance Storytelling Therapy workshops, it is also intended to analyze and verify the advantages of PWA in a real case.