Attribute selection with fuzzy decision reducts

Abstract

Rough set theory provides a methodology for data analysis based on the approximation of concepts in information systems. It revolves around the notion of discernibility: the ability to distinguish between objects, based on their attribute values. It allows to infer data dependencies that are useful in the fields of feature selection and decision model construction. In many cases, however, it is more natural, and more effective, to consider a gradual notion of discernibility. Therefore, within the context of fuzzy rough set theory, we present a generalization of the classical rough set framework for databased attribute selection and reduction using fuzzy tolerance relations. The paper unifies existing work in this direction, and introduces the concept of fuzzy decision reducts, dependent on an increasing attribute subset measure. Experimental results demonstrate the potential of fuzzy decision reducts to discover shorter attribute subsets, leading to decision models with a better coverage and with comparable, or even higher accuracy.

Fuzzy Covering based Rough Sets Revisited

“…For K x ∈ Cov(C) it holds that K x (x) = 1, so we have that Future work includes the study of fuzzy extensions of the loose pair of covering based rough set model, as well as the research of new fuzzy covering based rough set models. Moreover, we are interested in the applicability of fuzzy covering based rough sets in feature selection [18].…”

Improving the β-Precision and OWA Based Fuzzy Rough Set Models: Definitions, Properties and Robustness Analysis

“…Both models use aggregation operators instead of the inf-and sup-operator and preliminary work showed that they have interesting theoretical and practical assets. Unfortunately, they do not satisfy the inclusion property, which is required if we want the approximations to be on both sides of the set to be approximated, and is important for feature selection [8].…”

OWA-FRPS: A Prototype Selection Method Based on Ordered Weighted Average Fuzzy Rough Set Theory

“…The method that we develop is a wrapper. Although many researchers have focused on developing fuzzy rough feature selection [11] algorithms, there is not much literature on fuzzy rough PS yet. Nevertheless, fuzzy rough set theory [12] is a good tool to model noisy data.…”