Articles publicats (Enginyeria Informàtica i Disseny Digital)
Permanent URI for this collection
Browse
Recent Submissions
- ItemOpen AccessRhizomer-LLM: Natural language interface for semantic data exploration(Elsevier, 2026) Paslavska, D.; López-Gil, Juan-Miguel; Pereira, J.; Gil Iranzo, Rosa María; García González, RobertoExploring Semantic Web data typically requires expertise in SPARQL, the standard query language for Resource Description Framework (RDF) datasets. This technical barrier limits the accessibility of knowledge graphs to non-expert users. In this work, we present Rhizomer-LLM, an extension of the Rhizomer platform that integrates Large Language Models (LLMs) to enable natural language querying of semantic data. Users can input plain-text questions, which are automatically translated into SPARQL queries, executed against SPARQL endpoints via Apache Jena Fuseki, and visualized within Rhizomer’s interactive Zoom&Filter interface. The system architecture comprises an Angular frontend (RhizomerEye), a Java Spring backend (RhizomerAPI), and a Python service (ServerLLM) that handles LLM-based query generation. We evaluated the system using the BESDUI benchmark with four LLMs (Llama 3.3, DeepSeek R1, Gemini 2.5 Pro, and Gemini 2.5 Flash), demonstrating that this approach significantly improves accessibility to semantic data while maintaining high capability across diverse query types. Rhizomer-LLM is designed to help researchers and practitioners explore knowledge graphs using natural language, eliminating the need to learn complex query languages.
- ItemOpen AccessAtmósferas sonoras vernáculas y ritmos urbanos en barrios vulnerabilizados: una aproximación ritmoatmosférica al barrio de la Mariola de Lleida(Asociación Española de Geografía, 2026) Bonastra, Quim; Lega Lladós, FerranEn las últimas décadas, la investigación social ha mostrado un creciente interés por la dimensión sensible y afectiva de la experiencia urbana. Este estudio aborda la brecha existente entre los enfoques acústicos técnicos y los análisis culturales de la sonoridad, a través de un marco ritmoanalítico para comprender las atmósferas sonoras de barrios vulnerables. El objetivo es examinar cómo los sonidos cotidianos co-producen prácticas sociales, ritmos espaciales y sentidos de lugar en el barrio de la Mariola (Lleida). Para ello se combinaron mediciones acústicas en decibelios y registros de paisajes sonoros realizados entre 2023 y 2025, atendiendo a variaciones estacionales y horarias. Los resultados revelan una marcada polirritmia urbana: la coexistencia de ciclos de silencio matinal, clímax acústico vespertino y la centralidad del canto de los pájaros de competición como generador atmosférico vernáculo. La entrada en vigor de la Ley 7/2023 sobre bienestar animal provocó una disrupción sonora y afectiva, transformando un paisaje eurítmico en un escenario de arritmia y pérdida comunitaria. El artículo demuestra que la escucha situada permite reconocer la densidad afectiva de lo sonoro como elemento constitutivo de la vida.
- ItemRestrictedEvolving sex differences in suicide mortality by age in Spain, 2000–2023: a nationwide trend analysis(BMJ Publishing Group, 2026-01-01) Irigoyen-Otiñano, Maria; López-Molina, Gemma; Florensa Cazorla, Dídac; Gil-Conesa, Mario; Oliva, Vicenzo; Llorca-Bofí, VicentIntroduction: Sex disparities in suicide mortality persist worldwide, with male rates 2–4 times higher than female rates, but age-specific patterns over time remain insufficiently understood. Aim: To examine long-term trends in suicide mortality in Spain (2000–2023), focusing on sex-specific and age-specific differences and the evolution of the male-to-female suicide rate ratio. Methods: We conducted a longitudinal study using national mortality data from the Spanish Statistical Office. Crude and age-adjusted suicide rates were calculated by sex and age group. Joinpoint regression assessed temporal trends and male-to-female rate ratios across 15 age groups. Results: The overall age-adjusted rate showed no significant change (average annual percent change: –0.4%), with a significant decrease among men (–0.7%) and stable overall rates among women, while increases were observed in selected female age groups. Rates increased with age, particularly among men aged 70+. The male-to-female ratio widened with age, reaching 10.74 in the 95+ group, but narrowed over the study period in four age groups (15–29, 50–54, 75–79, 80–84), mainly due to declining male rates. Conclusion: Despite modest changes over time, marked age-related sex disparities persist. Older men remain the highest-risk group, while increases among women in midlife warrant targeted prevention.
- ItemRestrictedChronic obstructive pulmonary disease screening using time-frequency features of self-recorded respiratory sounds(Oxford University Press, 2025) Tena del Pozo, Alberto; Juez-Garcia, Ivan; D. Benítez, Iván; Clariá, Francesc; González, Jessica; De Batlle, Jordi; Solsona, FrancescObjectives: Chronic obstructive pulmonary disease (COPD) is the third leading cause of death worldwide, with up to 70% of cases remaining undiagnosed. This paper proposes a COPD screening tool based on time–frequency representation features of self-recorded respiratory sounds. Materials and Methods: Respiratory sound samples (breath and cough sounds) were extracted from COPD and asymptomatic non-COPD volunteers using a large, scientific-purpose database. We analyzed 39 time–frequency representation features of breath and cough sounds, combined with age, sex, and smoking status, using Autoencoder neural networks and random forest (RF) algorithms. We compared the performance of different breath and cough RF models built to detect COPD: one based exclusively on sound features, one based exclusively on sociodemographic characteristics, and one based on sound features and sociodemographic characteristics. Results: Models including breathing features outperformed models exclusively based on sociodemographic characteristics. Specifically, the model combining sociodemographic characteristics and breathing features achieved an area under the curve (AUC), accuracy, sensitivity, and specificity of 0.901, 0.836, 0.871, and 0.761, respectively, in the test set, representing a substantial increase in AUC when compared to the model based exclusively on sociodemographic characteristics (0.901 vs 0.818). Discussion: Our results suggest that a lightweight collection of the time–frequency representation features of self-recorded beathing sounds could effectively improve the predictive performance of COPD screening or case-finding questionnaires. Conclusion: COPD screening through self-recorded breathing sounds could be easily integrated as a low-cost first step in case-finding programs, potentially contributing to mitigate COPD underdiagnosis.
- ItemRestrictedDeep Learning for Detecting Abandoned Dogs(UNIR. Universidad en internet, 2026) Akinwehinmi, Oluwakemi; Tena del Pozo, Alberto; Mora, Francisco Javier; Solsona, Francesc; Del Amo, Pedro ArnauThis research paper presents a methodology consisting of an algorithm and a workflow for finding abandoned dogs in natural surroundings. We propose a temporal-contextual methodology for identifying abandoned pets in public areas. This involves employing a temporal rule alongside context-sensitive object identification, whereby dog bounding boxes are deliberately expanded to encompass proximate visual indicators suggestive of abandonment. The proposed approach uses object detection techniques, trajectory analysis, and image segmentation to quickly differentiate between abandoned and owned dogs. The research addresses key challenges, such as data scarcity and the complexity of distinguishing between abandoned and accompanied dogs. To address the issue of sufficient and adequate training corpus to identify an abandoned dog, the research article employs single-channel image augmentation methods that improve model recall and precision by 4%. Several object detection algorithms were evaluated, and our findings indicate that single-stage detectors like YOLOv8 achieved a better trade-off between classification performance and speed for detecting abandoned dogs compared to multi-stage detectors like Faster R-CNN, reaching a mean average precision up to 86% and an inference time of 0.3 ms per frame. This research study contributes to animal welfare, biodiversity conservation, and public safety by providing a scalable solution for monitoring abandoned animals in diverse environments. The findings demonstrate the impact of object detection techniques on improving the generalization of deep learning models for real-world applications.
