https://hdl.handle.net/10481/13885 2026-04-20T07:48:29Z 2026-04-20T07:48:29Z Vargas Pérez, Víctor Alejandro Chica Serrano, Manuel Cordón García, Óscar https://hdl.handle.net/10481/112667 2026-04-08T06:47:23Z

IPOP-CMA-ES and the Influence of Different Deviation Measures for Agent-Based Model Calibration Vargas Pérez, Víctor Alejandro; Chica Serrano, Manuel; Cordón García, Óscar Calibration is a crucial task on building valid models before exploiting their results. This process consists of adjusting the model parameters in order to obtain the desired outputs. Automatic calibration can be performed by using an optimization algorithm and a fitness function, which involves a deviation measure to compare the time series coming from the model. In this paper, we apply a memetic IPOP-CMA-ES for the calibration of an agent-based model and we study the effect of different deviation measures in this calibration problem. Classical metrics calculate the mean point-to-point error, but we also propose using an extension of dynamic time warping, which considers trend series evolution. In order to determine if calibrating with an specific metric leads to better solutions, we carry out an exhaustive experimentation by including statistical tests, analysis on the values of the calibrated parameters, and qualitative results. Our results show IPOP-CMA-ES obtains better performance than a genetic algorithm. In addition, MAE, MAPE and Soft-DTW are the metrics which report best results, although we get a similar behavior for all of them. This work is supported by the Spanish Agencia Estatal de Investigación, the Andalusian Government, the University of Granada, European Regional Development Funds (ERDF) under grants EXASOCO (PGC2018-101216-B-I00), SIMARK (P18-TP-4475), and AIMAR (A-TIC-284-UGR18) as well as by the Program "Becas de Iniciación UGR - Banco Santander".

Marín Ruiz, Nicolás Sánchez Fernández, Daniel Rivas Gervilla, Gustavo https://hdl.handle.net/10481/111082 2026-02-20T09:00:07Z

An Approximation to Context-Aware Size Modeling for Referring Expression Generation Marín Ruiz, Nicolás; Sánchez Fernández, Daniel; Rivas Gervilla, Gustavo In this paper we describe a methodology for modeling context-dependent fuzzy size categories like "small" and "large". We consider in this work that the context is fixed by a collection of crisp size values, so that the relativity in the definition of the categories is related to the distances between sizes in the context. Modeling visual concepts like those related to size is a key point, for instance, in the generation of referring expressions (conjunctions of properties) identifying objects in a certain visual scene. Taking context into account in the fuzzy modeling process is crucial in order to get human-like results. We illustrate our approach with several examples, comparing the results with other usual approaches to size category modeling.

Marín Ruiz, Nicolás Sánchez Fernández, Daniel Rivas Gervilla, Gustavo https://hdl.handle.net/10481/111072 2026-02-20T09:00:07Z

Fuzzy Group By Queries via RL-Instances in Conventional RDBMS Marín Ruiz, Nicolás; Sánchez Fernández, Daniel; Rivas Gervilla, Gustavo SQL group by queries allow grouping tuples in relational databases and computing aggregated values from the obtained sets of tuples, which makes them a particularly useful tool for data analysis. In this paper, we propose a way to solve this type of queries in a fuzzy environment, using fuzzy partitions of the domain of attributes according to Ruspini. The resolution of the query is carried out through the use of RL-instances and concepts from the theory of Representations by Levels, particularly a new method to derive RL-partitions from fuzzy partitions. The approach can be easily used in conventional systems through well-known SQL patterns.

Marín Ruiz, Nicolás Rivas Gervilla, Gustavo Sánchez Fernández, Daniel https://hdl.handle.net/10481/111038 2026-02-20T09:00:07Z

Referring under Uncertainty Marín Ruiz, Nicolás; Rivas Gervilla, Gustavo; Sánchez Fernández, Daniel In this paper, we study the referring expression generation problem (REG) when the available information about the properties of objects is uncertain, in the sense that we are not sure about the actual properties an object has. We formalize the problem by extending the conventional REG framework through the use of possibility distributions, represented by fuzzy sets. We show the potential benefits of this proposal in the assessment of the referential success for referring expressions. This approach opens a new line of research in the application of fuzzy sets to the REG problem, complementary to those approaches that use fuzzy sets as a suitable bridge between language and raw data. This work has been partially supported by the Spanish Government and the European Regional Development Fund (Fondo Europeo de Desarrollo Regional - FEDER) under projects TIN2014-58227-P Descripción lingüística de información visual mediante técnicas de minería de datos y computación flexible and project PGC2018-096156-B-I00. This work has also been partially supported by the Spanish Ministry of Education, Culture and Sports grant FPU16/05199

Cano Utrera, Andrés Gómez Olmedo, Manuel Moral Callejón, Serafín https://hdl.handle.net/10481/109913 2026-01-19T13:25:28Z

Learning sets of bayesian networks Cano Utrera, Andrés; Gómez Olmedo, Manuel; Moral Callejón, Serafín Este trabajo analiza el problema del aprendizaje de una red credal generalizada (un conjunto de redes bayesianas) a partir de un conjunto de datos. Se basa en el uso de la puntuación BDEu y calcula todas las redes con una puntuación superior a un factor predeterminado del óptimo. Para evitar el problema de determinar el tamaño muestral equivalente (ESS), el enfoque también considera la posibilidad de un ESS indeterminado. Aunque el resultado final es un conjunto de redes bayesianas, el trabajo también estudia el problema de seleccionar una única red con algunos procedimientos alternativos. Por último, se llevan a cabo algunos experimentos preliminares con tres redes pequeñas. This paper considers the problem of learning a generalized credal network (a set of Bayesian networks) from a dataset. It is based on using the BDEu score and computes all the networks with score above a predetermined factor of the optimal one. To avoid the problem of deter- mining the equivalent sample size (ESS), the approach also considers the possibility of an undetermined ESS. Even if the final result is a set of Bayesian networks, the paper also studies the problem of selecting a sin- gle network with some alternative procedures. Finally, some preliminary experiments are carried out with three small networks.