Soft Computing: Theory Innovations and Problem-Solving Benefits

Permanent URI for this collectionhttps://hdl.handle.net/10125/112440

Browse

Recent Submissions

Now showing 1 - 6 of 6
  • Thumbnail Image
    Item type: Item ,
    Analyzing Protocols of Information Granularity Allocation to Compute Missing Values in Intuitionistic Reciprocal Preference Relations
    (2026-01-06) González-Quesada, Juan Carlos; Díaz Jiménez, David; Perez Gálvez, Ignacio; Herrera-Viedma, Enrique; Cabrerizo, Francisco
    In the field of Granular Computing, a fundamental approach focuses on the management of incomplete intuitionistic reciprocal preference relations through the allocation of information granularity and the execution of a corresponding optimization procedure. It facilitates the transformation of numerical models into their granular counterparts, thereby providing a more accurate representation of reality to recover missing information. In decision-making contexts that involve intuitionistic reciprocal preference relations, this approach has proven essential in advancing procedures for estimating incomplete information. Nevertheless, while several protocols for information granularity allocation have been proposed, only one has been actively implemented thus far: the protocol based on a uniform and symmetric allocation of information granularity. To address this limitation, the objective of this study is to assess the effectiveness of existing protocols for allocating information granularity in the estimation of missing values in incomplete intuitionistic reciprocal preference relations. Numerical tests are included to demonstrate the efficacy of the protocols.
  • Thumbnail Image
    Item type: Item ,
    Emotions in Collective Risk Dilemmas Using Fuzzy Linguistic Rules
    (2026-01-06) Chica, Manuel; Cordon, Oscar; Abbass, Hussein
    The goal of this paper is to propose how to model emotions in climate change policies through evolutionary game theory (i.e., collective risk dilemmas) using fuzzy linguistic rules. We will first study evolutionary game theory models for exploring climate change policies and decisions to enhance cooperation between the policy players. Using computational agent-based simulations, we will build a framework where we incorporate and extend evolutionary game models with players' emotions in the game dynamics. These emotions, modeled by fuzzy linguistic rules, are a way for players to reconsider their strategies when playing the game. Results show how these rules representing emotions can better control defection in the collective-risk dilemmas while modeling a more realistic way of dealing with players' features.
  • Thumbnail Image
    Item type: Item ,
    Federated Learning for Brain Tumor Classification from MRI: A Comparison of MLP and ConvNeXt Approaches under IID and Non-IID Data Scenarios
    (2026-01-06) Miglionico, Giustino; Fazzolari, Michela; Ducange, Pietro; Marcelloni, Francesco; Ruffini, Fabrizio
    Federated Learning (FL) effectively addresses privacy concerns in medical imaging by enabling collaborative model training without sharing sensitive patient data. This paper compares two neural network approaches for brain tumor classification (BTC) from magnetic resonance imaging (MRI) in a federated learning setting. Both models operate on regions of interest (ROIs) extracted from brain MRI scans. The first is a lightweight multilayer perceptron (MLP) that classifies ROIs based on radiomic features extracted from them. The second is a deep learning (DL) approach based on the ConvNeXt architecture, which performs classification directly on the ROI images. Two experimental scenarios are considered: a balanced (IID) and an unbalanced (non-IID) distribution of data among federated clients. Results show that the radiomics-based MLP achieves performance comparable to the more complex ConvNeXt model, while requiring significantly lower computational resources. Moreover, federated learning consistently outperforms isolated local training, particularly under non-IID conditions, emphasizing its potential for clinical deployment.
  • Thumbnail Image
    Item type: Item ,
    Embracing Uncertainty in Human Activity Recognition: A Fuzzy Logic Framework for Interpretable and Context-aware Reasoning
    (2026-01-06) Díaz Jiménez, David; Martínez-Cruz, Carmen; Gaitán-Guerrero, Juan F.; López Ruiz, José Luis; Albín, Antonio P.; Espinilla, Macarena
    Human Activity Recognition (HAR) in real-world environments is inherently uncertain — shaped by ambiguous sensor signals, behavioral variability, and contextual dynamics that challenge traditional machine learning approaches. Despite growing interest in robustness and explainability, most current systems still treat uncertainty as a nuisance to be minimized rather than a structural feature to be modeled. This paper proposes a conceptual shift: positioning fuzzy logic as the core paradigm for designing HAR systems that are interpretable, adaptive, and uncertainty-aware. We present a theoretical framework in which fuzzy reasoning is integrated throughout the HAR pipeline— from sensor abstraction and context modeling to soft, interpretable activity inference and natural language explanations. By framing uncertainty as a representational and inferential asset, rather than a limitation, our approach enables systems that align more closely with the complexity of human behavior and the demands of human-centered AI. The framework is modular, extensible, and designed for transparency— making it suitable for long-term deployment in smart environments, particularly in domains like elderly care, remote monitoring, and assistive technologies. This work contributes a structured foundation for building next-generation HAR systems that move beyond black-box classification, supporting ethical, explainable, and context-sensitive activity recognition.
  • Thumbnail Image
    Item type: Item ,
    Enhancing Group Decision-Making through Large Language Models for Semantic Filtering of Expert Opinions
    (2026-01-06) Trillo, José Ramón; García-Cabello, Julia; Tapia, Juan Miguel; Alonso, Sergio; Morente-Molinera, Juan Antonio
    In modern decision-making contexts involving multiple experts, the use of natural language to express opinions introduces considerable difficulties for traditional analytical methods. With the growth of online collaboration, expert evaluations have become increasingly diverse in both format and scale, especially when communicated through unstructured textual comments. While conventional Group Decision-Making techniques can handle varied evaluation metrics, they cannot often fully interpret the semantic richness of human language. This work presents a novel Group Decision-Making framework that leverages a Large Language Model to semantically analyse and filter expert commentary, specifically DeepSeek LLM, identifying and removing remarks that may negatively influence others. Unlike standard sentiment analysis approaches based on fixed lexical rules, the Large Language Model captures subtle linguistic features such as tone, context, and implied meaning, allowing for a more accurate extraction of genuine expert preferences. The filtered and interpreted inputs are then aggregated to produce a consensus that better reflects the true collective judgment. This integration of advanced natural language processing enhances the interpretability, fairness, and reliability of group decisions in complex and dynamic environments. This constitutes a departure from traditional sentiment analysis approaches, as the LLM is used not only for interpretation but also for active semantic filtering of expert discourse.
  • Thumbnail Image
    Item type: Item ,
    Introduction to the Minitrack on Soft Computing: Theory Innovations and Problem-Solving Benefits
    (2026-01-06) Herrera-Viedma, Enrique; Cabrerizo, Francisco; Perez Gálvez, Ignacio