Javier Gonzalez

Javier Gonzalez

Auto-generated profile, based on publicly available information

Javier Gonzalez

Syst. & Eng. Dept., Univ. of Malaga, Malaga

Research Focus & Metrics

Academic Career Length

28Years (1999-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 1999 in Animal Research
Verified
Journal Article
Open access
PDF file
María Cecilia CajarvilleMaría Cecilia Cajarville
María Cecilia Cajarville
JGJavier González
Javier González
JRJosé Luis Repetto
José Luis Repetto
CRCarlos Alberto Rodríguez
Carlos Alberto Rodríguez
AMAdela Martínez
Adela Martínez
The nutritive value of cardoon (Cynara cardunculu.s L.) green forage (GF), dry leaves (DL) and straw (S) was studied from chemical composition analyses and from voluntary intake, digestibility and rumen degradability estimations obtained using adult wethers. Aptitude for being ensiled according to the wilting period was also evaluated making silages with fresh forage and after 24 and 94 hours of wilting. Green forage had a high nutritive value, as shown by low levels of fibre and lignin and its very high digestibility (86 % for OM). Nevertheless, its voluntary intake was relatively low (64 g DM.kg!' L W O .7 5), although very variable among animals. Fermentation characteristics of the GF silages were good and wilting reduced silage losses and improved the quality of silages. Nutritive value of DL was high, but limited only by a moderate crude protein content. Supplemented with urea, it had an OM digestibility similar to a medium-good quality hay (65.9 % average) and a voluntary intake equivalent to a good quality grass (75.9 g DM.kg!' LW(7 5). Finally, S was a typical fibrous and low protein forage, but when supplemented with urea, it had an intake and a digestibility higher than those of cereal straws. Variations in digestibility for OM and energy among the different non-preserved Cynara cardunculus forages showed close relationships with fibre fractions , especially with the acid detergent lignin content. " (© Elsevier / Inra). Cardoon / degradability / digestibility / intake / silage Résumé — Valeur nutritive du fourrage vert et des sous-produits de culture de Cynara car-dunculus. La valeur nutritive du fourrage vert, des feuilles sèches et de « la paille » de Cardon (Cynara cardunculus L.) a été étudiée à partir des analyses de composition chimique, et des estimations de digestibilité, de dégradabilité et d'ingestion volontaire chez le mouton. La valeur énergétique des

2 May 2008 in IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics)
Verified
Journal Article
Closed access
CGCipriano Galindo
Cipriano Galindo
Juan-Antonio Fernández-MadrigalJuan-Antonio Fernández-Madrigal
Juan-Antonio Fernández-Madrigal
JGJavier Gonzalez
Javier Gonzalez
To date, no solution has been proposed to human-machine interactive task planning that deals simultaneously with two important issues: 1) the capability of processing large amounts of information in planning (as it is needed in any real application) and 2) being efficient in human-machine communication (a proper set of symbols for human-machine interaction may not be suitable for efficient automatic planning and vice versa). In this paper, we formalize a symbolic model of the environment to solve these issues in a natural form through a human-inspired mechanism that structures knowledge in multiple hierarchies. Planning with a hierarchical model may be efficient even in cases where the lack of hierarchical information would make it intractable. However, in addition, our multihierarchical model is able to use the symbols that are most familiar to each human user for interaction, thus achieving efficiency in human-machine communication without compromising the task-planning performance. We formalize here a general interactive task-planning process which is then particularized to be applied to a mobile robotic application. The suitability of our approach has been demonstrated with examples and experiments.

2 January 2025 in Informática y Sistemas: Revista de Tecnologías de la Informática y las Comunicaciones
Verified
Journal Article
Open access
KCKevin Alexander Mendoza Campoverde
Kevin Alexander Mendoza Campoverde
JGJavier Valentin Hurtado Gonzalez
Javier Valentin Hurtado Gonzalez
RRRodrigo Fernando Morocho Román
Rodrigo Fernando Morocho Román
WRWilmer Rivas
Wilmer Rivas
El presente trabajo muestra una comparación entre dos modelos de inteligencia artificial para la detección de lenguaje agresivo en redes sociales entre un modelo tradicional de clasificación de texto y un modelo basado en redes neuronales profundas. Se utilizaron dos enfoques principales: regresión logística utilizando vector TF-IDF y un modelo basado en BERT adaptado para procesamiento de lenguaje natural. En cuanto a la metodología se aplicó CRISP-DM, abordando desde la preparación de los datos hasta la parte final que es la evaluación de los modelos. Se presentaron balances en el conjunto de datos, el cual se corrigió usando la técnica SMOTE. La evaluación de modelos nos demostró que BERT alcanzó mejores métricas de rendimiento con una medida F1 promedio de 0.93 en comparación a la regresión logística que presentó un 0.83. Las métricas junto con la revisión de errores de clasificación ayudaron a observar de forma más clara en qué aspectos cada enfoque presentaba fortalezas o mostraba limitaciones. En síntesis, los resultados obtenidos manifiestan que BERT ofrece ventajas importantes para la tarea de moderación de contenido en redes sociales y además se pudo confirmar que el preprocesamiento adecuado y el balanceo de los datos son factores clave para mejorar el rendimiento en problemas relacionados con la clasificación de texto.