Boosting fuzzy rules in classification problems under single-winner inference

Sánchez, Luciano; Otero, José

doi:10.1002/int.20236

Cited by 39 publications

(18 citation statements)

References 20 publications

(23 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In previous works, we have used a genetic algorithm for finding the function f, and tested the preceding search scheme for extracting fuzzy rules from crisp data in both classifiers and regression models (del Jesus et al 2004;Otero and Sanchez 2006;Sánchez and Otero 2007). The method is accurate and has been shown to be as fast as some ad-hoc learning methods (Sánchez and Otero 2004).…”

Section: Backfitting In Crisp Datasetsmentioning

confidence: 97%

Obtaining linguistic fuzzy rule-based regression models from imprecise data with multiobjective genetic algorithms

2008

Self Cite

View full text Add to dashboard Cite

Backfitting of fuzzy rules is an Iterative RuleLearning technique for obtaining the knowledge base of a fuzzy rule-based system in regression problems. It consists in fitting one fuzzy rule to the data, and replacing the whole training set by the residual of the approximation. The obtained rule is added to the knowledge base, and the process is repeated until the residual is zero, or near zero. Such a design has been extended to imprecise data for which the observation error is small. Nevertheless, when this error is moderate or high, the learning can stop early. In this kind of algorithms, the specificity of the residual might decrease when a new rule is added. There may happen that the residual grows so wide that it covers the value zero for all points (thus the algorithm stops), but we have not yet extracted all the information available in the dataset. Focusing on this problem, this paper is about datasets with medium to high discrepancies between the observed and the actual values of the variables, such as those containing missing values and coarsely discretized data. We will show that the quality of the iterative learning degrades in this kind of problems, because it does not make full use of all the available information. As an alternative to sequentially obtaining rules, we propose a new multiobjective Genetic Cooperative Competitive Learning (GCCL) algorithm. In our approach, each individual in the population codifies one rule, which competes in the population in terms of maximum coverage and fitting, while the individuals in the population cooperate to form the knowledge base.

show abstract

Section: Backfitting In Crisp Datasetsmentioning

confidence: 97%

Obtaining linguistic fuzzy rule-based regression models from imprecise data with multiobjective genetic algorithms

2008

Self Cite

View full text Add to dashboard Cite

show abstract

“…Carse and Pipe's special issue [13] includes papers focused on the multiobjective evolutionary learning [37], boosting [63] and evolutionary adaptive inference systems [5]. Casillas et al's special issue [19] is focused on the trade-off between interpretability and accuracy, collecting papers that proposed different GFSs for tackling this problem: with multiobjective approaches [47,39], and optimizing the definition for the linguistic variables [4,11].…”

Section: Genetic Fuzzy Systemsmentioning

confidence: 99%

GP-COACH: Genetic Programming-based learning of COmpact and ACcurate fuzzy rule-based classification systems for High-dimensional problems

Berlanga

Rivera

Jesus

et al. 2010

Information Sciences

125

View full text Add to dashboard Cite

“…This scenario also includes boosting of fuzzy rules 16,58 . In both schemes, a single rule is created step by step and examples are then weighted or removed according to the coverage of the current rule base at that point.…”

Section: How Can It Be Designed?mentioning

confidence: 99%

A View on Fuzzy Systems for Big Data: Progress and Opportunities

Fernández¹,

Carmona²,

Jesus³

et al. 2016

IJCIS

View full text Add to dashboard Cite

Currently, we are witnessing a growing trend in the study and application of problems in the framework of Big Data. This is mainly due to the great advantages which come from the knowledge extraction from a high volume of information. For this reason, we observe a migration of the standard Data Mining systems towards a new functional paradigm that allows at working with Big Data. By means of the MapReduce model and its different extensions, scalability can be successfully addressed, while maintaining a good fault tolerance during the execution of the algorithms. Among the different approaches used in Data Mining, those models based on fuzzy systems stand out for many applications. Among their advantages, we must stress the use of a representation close to the natural language. Additionally, they use an inference model that allows a good adaptation to different scenarios, especially those with a given degree of uncertainty. Despite the success of this type of systems, their migration to the Big Data environment in the different learning areas is at a preliminary stage yet. In this paper, we will carry out an overview of the main existing proposals on the topic, analyzing the design of these models. Additionally, we will discuss those problems related to the data distribution and parallelization of the current algorithms, and also its relationship with the fuzzy representation of the information. Finally, we will provide our view on the expectations for the future in this framework according to the design of those methods based on fuzzy sets, as well as the open challenges on the topic.

show abstract

Boosting fuzzy rules in classification problems under single-winner inference

Cited by 39 publications

References 20 publications

Obtaining linguistic fuzzy rule-based regression models from imprecise data with multiobjective genetic algorithms

Obtaining linguistic fuzzy rule-based regression models from imprecise data with multiobjective genetic algorithms

GP-COACH: Genetic Programming-based learning of COmpact and ACcurate fuzzy rule-based classification systems for High-dimensional problems

A View on Fuzzy Systems for Big Data: Progress and Opportunities

Contact Info

Product

Resources

About