http://hdl.handle.net/11531/44 Artículos de revista, capítulos de libro y contribuciones en congresos publicadas. Sat, 13 Jun 2026 11:47:31 GMT 2026-06-13T11:47:31Z http://hdl.handle.net/11531/110698 SemTab: A Hybrid Framework for Semantic Feature Generation on Tabular Data Chen, Olivia; Chou, Kara; Nagpal, Rashmi; Palacios Hielscher, Rafael; Gupta, Amar Machine learning models on tabular datasets often struggle to understand the context between features, which can limit their accuracy. We propose SemTab, a hybrid framework for generating semantic features that utilizes an open-source Large Language Model (LLM). We evaluated our framework using three benchmark datasets: Adult Income, German Credit, and Bank Marketing. We compared its performance against several off-the-shelf LLMs. The results show that SemTab achieved the highest accuracy across all the classification tasks. For instance, on the Bank Marketing dataset, SemTab achieved an accuracy of 8 0%, which is approximately 2 0% improvement over the baseline models. This work highlights that a hybrid architecture is a practical approach for applying language models to structured tabular data, yielding accurate and interpretable results for various downstream tasks.; Machine learning models on tabular datasets often struggle to understand the context between features, which can limit their accuracy. We propose SemTab, a hybrid framework for generating semantic features that utilizes an open-source Large Language Model (LLM). We evaluated our framework using three benchmark datasets: Adult Income, German Credit, and Bank Marketing. We compared its performance against several off-the-shelf LLMs. The results show that SemTab achieved the highest accuracy across all the classification tasks. For instance, on the Bank Marketing dataset, SemTab achieved an accuracy of 8 0%, which is approximately 2 0% improvement over the baseline models. This work highlights that a hybrid architecture is a practical approach for applying language models to structured tabular data, yielding accurate and interpretable results for various downstream tasks. Capítulos en libros Tue, 26 May 2026 00:00:00 GMT http://hdl.handle.net/11531/110698 2026-05-26T00:00:00Z http://hdl.handle.net/11531/110697 A generalized kinetic model for anaerobic co-digestion of agro-industrial biomass mixtures improved through pretreatments or additives Victoria Rodríguez, Javier; Morales Polo, Carlos; Cledera Castro, María del Mar Anaerobic co-digestion is increasingly recognised as a strategy to enhance biogas production from agro-industrial biomass by exploiting synergistic interactions among multiple organic substrates. However, most existing kinetic models remain case-specific and fail to provide reliable predictions when substrate ratios, pretreatments, or additives are modified, limiting their practical applicability. This study proposes a generalised combinatorial kinetic framework for n-substrate anaerobic co-digestion. The model integrates established formulations, including the Modified Gompertz and first-order models, into a structured interaction-based approach capable of explicitly representing multi-order substrate interactions without repetition. Interaction coefficients are weighted according to substrate contributions, ensuring consistency between mono-digestion and higher-order co-digestion systems. The framework further incorporates pretreatments and nanoparticle additives through multiplicative interaction terms, allowing performance enhancements to be modelled explicitly rather than indirectly absorbed into kinetic parameters. Although the combinatorial structure can manage up to 108 interaction parameters in a three-substrate Gompertz-based system, the methodology reduces the minimum experimental requirement to only 8 BMP tests by sequentially constructing higher-order interactions from lower-order combinations. This significantly decreases experimental effort while preserving scalability. While the model currently simplifies additive concentration effects and potential inhibitory phenomena, it provides a flexible and extensible structure capable of evolving with emerging research trends. Overall, the proposed framework addresses a critical gap in anaerobic co-digestion modelling and supports a more systematic design of agro-industrial biomass co-digestion systems for bioenergy production within circular bioeconomy strategies.; Anaerobic co-digestion is increasingly recognised as a strategy to enhance biogas production from agro-industrial biomass by exploiting synergistic interactions among multiple organic substrates. However, most existing kinetic models remain case-specific and fail to provide reliable predictions when substrate ratios, pretreatments, or additives are modified, limiting their practical applicability. This study proposes a generalised combinatorial kinetic framework for n-substrate anaerobic co-digestion. The model integrates established formulations, including the Modified Gompertz and first-order models, into a structured interaction-based approach capable of explicitly representing multi-order substrate interactions without repetition. Interaction coefficients are weighted according to substrate contributions, ensuring consistency between mono-digestion and higher-order co-digestion systems. The framework further incorporates pretreatments and nanoparticle additives through multiplicative interaction terms, allowing performance enhancements to be modelled explicitly rather than indirectly absorbed into kinetic parameters. Although the combinatorial structure can manage up to 108 interaction parameters in a three-substrate Gompertz-based system, the methodology reduces the minimum experimental requirement to only 8 BMP tests by sequentially constructing higher-order interactions from lower-order combinations. This significantly decreases experimental effort while preserving scalability. While the model currently simplifies additive concentration effects and potential inhibitory phenomena, it provides a flexible and extensible structure capable of evolving with emerging research trends. Overall, the proposed framework addresses a critical gap in anaerobic co-digestion modelling and supports a more systematic design of agro-industrial biomass co-digestion systems for bioenergy production within circular bioeconomy strategies. Artículos en revistas Wed, 01 Jul 2026 00:00:00 GMT http://hdl.handle.net/11531/110697 2026-07-01T00:00:00Z http://hdl.handle.net/11531/110696 Analyzing the fictional worlds of Pixar with an eye on digital humanities Candel Bormann, Daniel; Giuliani Pedraza, Marta; Madarova, Slavka; Rubio Cáceres, Paula; Ruiz Sanz, Marta; Troyano Fernández, María Victoria; Treija, Kristīne El artículo busca cerrar la brecha entre las humanidades digitales y la crítica tradicional mediante una combinación de lectura distante y lectura cercana. Para ello construye visualizaciones, abre perspectivas para la codificación y analiza un corpus amplio formado por las películas de Pixar hasta The Good Dinosaur. El estudio emplea una herramienta de análisis orientada al examen de mundos ficcionales, situada entre el valor interpretativo y el método. Tras presentar dicha herramienta, los personajes principales de Pixar se ubican dentro de una visualización espaciotemporal y se interpreta su distribución semántica. El trabajo muestra cómo los métodos digitales pueden apoyar la interpretación crítica de narrativas audiovisuales complejas.; The present article tries to close the gap between the digital humanities and traditional criticism. It does so by moving between distant and close reading, constructing visualizations, opening perspectives for coding, and analyzing a large corpus: the Pixar movies up to The Good Dinosaur. This is done with the help of a tool of analysis which studies fictional worlds and is wedged between value and method. After introducing the tool, the main Pixar characters are placed within a spatiotemporal visualization of the tool and their semantic distribution is interpreted. The article shows how digital methods can support critical interpretation of complex audiovisual narratives and contribute to the study of fictional worlds. Artículos en revistas Tue, 25 Jul 2017 00:00:00 GMT http://hdl.handle.net/11531/110696 2017-07-25T00:00:00Z http://hdl.handle.net/11531/110675 Aproximación al canon 702. Equidad y caridad evangélica Rodríguez Fernández, Miguel El capítulo analiza el canon 702 del Código de Derecho Canónico, centrado en la situación de los miembros que abandonan o son expulsados de un instituto religioso. Aunque el canon niega el derecho a exigir compensaciones económicas por los servicios prestados durante la vida consagrada, el autor subraya que esta norma debe interpretarse junto con la obligación de observar la equidad y la caridad evangélica. El texto defiende que los institutos religiosos deben evitar respuestas meramente formalistas y establecer criterios claros de ayuda, teniendo en cuenta edad, salud, formación, duración de la vida religiosa, situación económica y posibilidades reales de reinserción. También propone mecanismos de mediación, acompañamiento y apoyo institucional que protejan la dignidad de la persona y refuercen una justicia eclesial inspirada en el servicio.; The chapter examines canon 702 of the Code of Canon Law, focusing on members who leave or are dismissed from a religious institute. Although the canon denies former members the right to claim financial compensation for services provided during religious life, the author stresses that this provision must be read alongside the institute’s duty to act with equity and evangelical charity. The text argues that religious institutes should avoid a purely formal or legalistic response and should establish clear criteria for assistance, considering age, health, education, length of religious profession, financial circumstances, and realistic prospects for social and professional reintegration. It also proposes mediation structures, pastoral support, and institutional accompaniment to safeguard personal dignity and promote an ecclesial model of justice rooted in service. Capítulos en libros Sat, 25 Apr 2026 00:00:00 GMT http://hdl.handle.net/11531/110675 2026-04-25T00:00:00Z