Fuzzy Rough Attribute Reduction for Categorical Data.

Changzhong Wang,Yan Wang,Mingwen Shao,Yuhua Qian,Degang Chen

DOI: https://doi.org/10.1109/tfuzz.2019.2949765

IF: 12.253

2020-01-01

IEEE Transactions on Fuzzy Systems

Abstract:Classical rough set theory is considered a useful tool for dealing with the uncertainty of categorical data. The major deficiency of this model is that the classical rough set model is sensitive to noise in classification learning due to the stringent condition of equivalence relation. Thus, a class of fuzzy similarity relations was introduced to describe the similarity between samples with categorical attributes. However, these kinds of similarity relations also have deficiencies when they are used in fuzzy rough computation. In this article, we propose a new fuzzy-rough-set model for categorical data by introducing a variable parameter to control the similarity of samples. This model employs the iterative computation strategy to define fuzzy rough approximations and dependence functions. It is proved that the proposed rough dependence function is monotonic. Finally, the proposed model is applied to the attribute reduction of categorical data. The experimental results indicate that the proposed model is more effective for categorical data than some existing algorithms.