DeustoTeka
DeustoTeka recoge la producción científica del personal docente e investigador de la Universidad de Deusto. Su propósito es reunir, archivar, preservar y aumentar la visibilidad en acceso abierto de los resultados de investigación.
Comunidades en DSpace
Selecciona una comunidad para explorar sus colecciones.
Envíos recientes
Exploring assessment measures including psychological aspects in adults with acquired Peripheral Facial Palsy (PFP): a scoping review
(BioMed Central Ltd, 2025-12-22) Polanco Fernández, Rocío; Penas, Patricia; Iraurgi Castillo, Ioseba
Introduction: Given the lack of specific assessment measures and consensus on how to evaluate the psychological aspects of acquired peripheral facial palsy, this study aimed to identify the existing instruments in current use, as well as the key psychological domains considered. A secondary objective was to identify mental health-related areas not addressed or under-represented by those instruments. Method: A scoping review was conducted to identify assessment measures, their underlying dimensions, and frequency of use. Findings were thematically synthesized across specific areas and grouped by consensus among the three authors to identify key domains for evaluating facial palsy. Results: 46 instruments were identified, also capturing social and quality-of-life aspects often closely linked to psychological factors. Their dimensions were classified into five generic areas: health-related quality of life perception, symptomatology (emotional and physical), appearance-related matters, intrapersonal and social skills, and care experience. Analysis of the items led to the identification of specific areas, which were classified by relevance into six broader psychosocial domains: (1) social functioning, (2) emotional symptomatology, (3) facial palsy- related QoL, (4) general health-related QoL, (5) psychological functioning, and (6) care experience-satisfaction. The FDI, FaCE, followed by the SF-36, and HADS emerged as the most used instruments. Conclusions: The results showed a lack of consensus due to the heterogeneity of the instruments retrieved for the evaluation of psychological outcomes. This review identified under-represented psychological areas, such as social and appearance-related anxiety, body self-perception, and experience of the care process which may warrant further evaluation. Several of the instruments retrieved were general rather than facial palsy-specific, which could not fully address the specific psychological needs of adults suffering with acquired peripheral facial palsy.
MRE-KDD+: an innovative multi-resolution, ensemble framework for supporting OLAM-based big data analytics over big data warehouses
(IGI Global, 2025) Cuzzocrea, Alfredo; García Bringas, Pablo
Big data settings are currently evolving from classical systems that focus on supporting advanced decision-support processes—as applied to many real-life scenarios, which are typically populated by distributed and heterogeneous data sources, such as conventional distributed data warehousing environments—to cooperative information systems. Different data formats contribute to define challenging big data systems, in which the main issue consists in supporting modern big data analytics involving massive amounts of data. As a consequence, a relevant research challenge is how to efficiently integrate, process, and mine such distributed knowledge, which composes the foundations of final big data analytics processes. Starting from these considerations, in this paper the authors propose an online analytical mining-based framework for supporting big data analytics, along with a formal model underlying this framework, called Multi-Resolution Ensemble-Based Model for Advanced Knowledge Discovery in Big Data Warehouses.
Large Language Models for structured task decomposition in Reinforcement Learning problems with sparse rewards
(Multidisciplinary Digital Publishing Institute (MDPI), 2025-10-22) Ruiz Gonzalez, Unai; Andrés Fernández, Alain; Ser Lorente, Javier del
Reinforcement learning (RL) agents face significant challenges in sparse-reward environments, as insufficient exploration of the state space can result in inefficient training or incomplete policy learning. To address this challenge, this work proposes a teacher–student framework for RL that leverages the inherent knowledge of large language models (LLMs) to decompose complex tasks into manageable subgoals. The capabilities of LLMs to comprehend problem structure and objectives, based on textual descriptions, can be harnessed to generate subgoals, similar to the guidance a human supervisor would provide. For this purpose, we introduce the following three subgoal types: positional, representation-based, and language-based. Moreover, we propose an LLM surrogate model to reduce computational overhead and demonstrate that the supervisor can be decoupled once the policy has been learned, further lowering computational costs. Under this framework, we evaluate the performance of three open-source LLMs (namely, Llama, DeepSeek, and Qwen). Furthermore, we assess our teacher–student framework on the MiniGrid benchmark—a collection of procedurally generated environments that demand generalization to previously unseen tasks. Experimental results indicate that our teacher–student framework facilitates more efficient learning and encourages enhanced exploration in complex tasks, resulting in faster training convergence and outperforming recent teacher–student methods designed for sparse-reward environments.
Enhanced short-term scheduling of underground mining activities using Tabu Search: a comparative analysis
(Universal Wiser Publisher, 2025-10-29) Álvarez Paredes, Luis; Hernández Guerra, Heber; Alberdi Celaya, Elisabete; Goti Elordi, Aitor
The planning of preparation and development activities in underground mining is essential to ensure efficiency and operational continuity. However, short-term scheduling of these tasks has received limited attention in literature. This study proposes a Cooperative Multi-start Tabu Search with Path-Relinking (CMTS-PR), which coordinates multiple tabu trajectories and intensifies them through path relinking to optimize the short-term scheduling of multiple underground heading works within a one-shift horizon. The problem is modeled as a flexible job-shop scheduling problem with unrelated parallel equipment and sequence dependent setup times. CMTS-PR is evaluated against a memetic algorithm, a Non-dominated Sorting Genetic Algorithm II, a single-trajectory Tabu Search (TS), a Constraint Programming (CP) model, and manual scheduling by an expert planner, across two panel caving case studies in Chile. The results show that CP yields mathematically optimal solutions but becomes computationally demanding, while manual scheduling ensures feasibility but underutilizes resources. In contrast, CMTS-PR produces operationally viable schedules. In case study CMTS-PR matched CP on equivalent fronts within 60 seconds, even under 10-60 minute transfer time variability. In case study 2, CMTS-PR increased equivalent fronts by 120% compared to manual planning and by 2.94% relative to CP, with lower runtime. Overall, CMTS-PR proves to be effective and computationally efficient, representing one of the first applications of a cooperative TS and path-relinking scheme to underground short-term scheduling, and providing practical tool for daily mine operations.
Generative and responsible AI in triadic Virtual Exchanges for tourism education
(Routledge, 2026-03-31) Aliperti, Giuseppe; Garza, Armida de la
While Generative AI is reshaping higher education, its role in tourism-focused Virtual Exchange remains underexplored. This study examines the impact of AI on student interaction, academic performance, and engagement in a pilot Virtual Exchange between universities in Spain and Ireland. Drawing on Conversation Theory, Virtual Exchange is conceptualized as a triadic process involving students and AI, moving from interaction to shared understanding and collaborative action. Using a mixed-methods approach with qualitative emphasis, the study explores how AI tools shaped the learning experience, highlighting both opportunities and challenges. Grounded in frameworks of women’s empowerment in tourism, internationalization of the curriculum, intercultural competence, and digital pedagogy, the findings deepen understanding of AI’s potential and limitations in Virtual Exchange contexts. The study also offers practical guidance for educators on developing AI literacy, promoting ethical awareness, and enhancing student collaboration in online international learning environments.