Novel Improvements on the Fuzzy-Rough QuickReduct Algorithm

Javad Rahimipour ANARAKI  Mahdi EFTEKHARI  Chang Wook AHN  

IEICE TRANSACTIONS on Information and Systems   Vol.E98-D   No.2   pp.453-456
Publication Date: 2015/02/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.2014EDL8099
Type of Manuscript: LETTER
Category: Pattern Recognition
fuzzy-rough set,  dependency degree,  feature selection,  fuzzy-rough quickreduct,  

Full Text: PDF(125.3KB)
>>Buy this Article

Feature Selection (FS) is widely used to resolve the problem of selecting a subset of information-rich features; Fuzzy-Rough QuickReduct (FRQR) is one of the most successful FS methods. This paper presents two variants of the FRQR algorithm in order to improve its performance: 1) Combining Fuzzy-Rough Dependency Degree with Correlation-based FS merit to deal with a dilemma situation in feature subset selection and 2) Hybridizing the newly proposed method with the threshold based FRQR. The effectiveness of the proposed approaches are proven over sixteen UCI datasets; smaller subsets of features and higher classification accuracies are achieved.