Pedro Villar

Generating the knowledge base of a fuzzy rule-based system by the genetic learning of the data base

A method is proposed to automatically learn the knowledge base by finding an appropiate data base by means of a genetic algorithm while using a simple generation method to derive the rule base. Our genetic process learns the number of linguistic terms per variable and the membership function parameters that define their semantics, while a rule base generation method learns the number of rules and their composition.

Analysis and guidelines to obtain a good uniform fuzzy partition granularity for fuzzy rule-based systems using simulated annealing☆

Abstract In this contribution, we will analyse the importance of the fuzzy partition granularity for the linguistic variables in the design of fuzzy rule-based systems (FRBSs). In order to put this into effect, we will study the FRBS behaviour considering uniform fuzzy partitions with the same number of labels for all the linguistic variables, and considering uniform fuzzy partitions with any number of labels for each linguistic variable. We will present a method based on Simulated Annealing (SA) in order to obtain a good uniform fuzzy partition granularity that improves the FRBS behaviour. It is an efficient granularity search method for finding a good number of labels per variable.

A genetic learning process for the scaling factors, granularity and contexts of the fuzzy rule-based system data base

Abstract In this contribution, we propose a new method to automatically learn the Knowledge Base of a Fuzzy Rule-Based System by finding an appropriate Data Base using a Genetic Algorithm and considering a simple generation method to derive the Rule Base. Our genetic process learns all the components of the Data Base (number of labels, working ranges and membership function shapes for each linguistic variable) using a non-linear scaling function to adapt the fuzzy partition contexts.

A multiobjective genetic algorithm for feature selection and granularity learning in fuzzy-rule based classification systems

We propose a new method to automatically learn the knowledge base of a fuzzy rule-based classification system (FRBCS) by selecting an adequate set of features and by finding an appropiate granularity for them. This process uses a multiobjective genetic algorithm and considers a simple generation method to derive the fuzzy classification rules.

A web tool to support decision making in the housing market using hesitant fuzzy linguistic term sets

Graphical abstractDisplay Omitted In this paper we present a linguistic multiple-expert multi-criteria decision making model and a web tool to support it, that is centred on the housing market. The web tool is integrated with the usual catalogue of resources for rental or for sale, enriched with the possibility of ranking a subset of properties according to the clients preferences and the internal knowledge associated to the properties. Usually the description of a property is quantitative, thought in our case we add qualitative information corresponding to assessments made by housing agents. These agents are considered experts in the market conditions.We apply the 2-tuple linguistic representation model to keep accuracy in the processes of Computing with Words and the hesitant fuzzy linguistic term sets to qualify in situations of uncertainty and hesitation in the assessments. The software helps the agents in the process of the elicitation of the linguistic expression based on the fuzzy linguistic approach and the use of context-free grammars, and the web clients in the decision of visiting a property.

Linguistic modeling with hierarchical systems of weighted linguistic rules

Recently, many different possibilities to extend the Linguistic Fuzzy Modeling have been considered in the specialized literature with the aim of introducing a trade-off between accuracy and interpretability. These approaches are not isolated and can be combined among them when they have complementary characteristics, such as the hierarchical linguistic rule learning and the weighted linguistic rule learning. In this paper, we propose the hybridization of both techniques to derive Hierarchical Systems of Weighted Linguistic Rules. To do so, an evolutionary optimization process jointly performing a rule selection and the rule weight derivation has been developed. The proposal has been tested with two real-world problems achieving good results.

FEATURE SELECTION AND GRANULARITY LEARNING IN GENETIC FUZZY RULE-BASED CLASSIFICATION SYSTEMS FOR HIGHLY IMBALANCED DATA-SETS

This paper proposes a Genetic Algorithm for jointly performing a feature selection and granularity learning for Fuzzy Rule-Based Classification Systems in the scenario of highly imbalanced data-sets. We refer to imbalanced data-sets when the class distribution is not uniform, a situation that it is present in many real application areas. The aim of this work is to get more compact models by selecting the adequate variables and adapting the number of fuzzy labels for each problem, improving the interpretability of the model. The experimental analysis is carried out over a wide range of highly imbalanced data-sets and uses the statistical tests suggested in the specialized literature.

A Multiobjective Genetic Learning Process for joint Feature Selection and Granularity and Contexts Learning in Fuzzy Rule-Based Classification Systems

In this contribution, we propose a genetic process to select an appropiate set of features in a Fuzzy Rule-Based Classification System (FRBCS) and to automatically learn the whole Data Base definition using a non linear scaling function to adapt the fuzzy partition contexts and determining an appropiate granularity for each of them. An ad-hoc data covering learning method is considered to obtain the Rule Base. The method uses a multiobjective genetic algorithm in order to obtain a good trade-off between accuracy and interpretability.

Multiple Crossover per Couple with Selection of the Two Best Offspring: An Experimental Study with the BLX-alpha Crossover Operator for Real-Coded Genetic Algorithms

In this paper, we propose a technique for the application of the crossover operator that generates multiple descendants from two parents and selects the two best offspring to replace the parents in the new population. In order to study the proposal, we present different instances based on the BLX-? crossover operator for real-coded genetic algorithms. In particular, we investigate the influence of the number of generated descendants in this operator, the number of evaluations, and the value for the parameter ?. Analyzing the experimentation that we have carried out, we can observe that it is possible, with multiple descendants, to achieve a suitable balance between the explorative properties associated with BLX-? and the high selective pressure associated to the selection of the two best descendants.

A Linguistic Multi-Criteria Decision Making Model Applied to the Integration of Education Questionnaires

We present a model made up of linguistic multi-criteria decision making processes to integrate the answers to heterogeneous questionnaires, based on a five-point Likert scale, into a unique form rooted in the widespread course experience questionnaire . The main advantage of having the resulting integrated questionnaire is that it can be incorporated into other course experience questionnaire surveys to make benchmarking among organizations. This model has been applied to integrate heterogeneous educational questionnaires at the University of Granada.