Abstract | ||
---|---|---|
Abstract The fuzzy set theory and the rough set theory are two distinct but complementary theories that deal with uncertainty in data. The salient features of both the theories are encompassed in the domain of the fuzzy rough set theory so as to cope with the problems of vagueness and indiscernibility in real world data. This hybrid theory has been found to be a potential tool for data mining, particularly useful for feature selection. Most of the existing approaches to fuzzy rough sets are based on fuzzy relations. In this paper, a new definition for fuzzy rough sets in an information system based on the divergence measure of fuzzy sets is introduced. The properties of the fuzzy rough approximations are explored. Moreover, an algorithm for feature selection using the proposed approximations is presented and experimented using real data sets. |
Year | DOI | Venue |
---|---|---|
2018 | 10.1016/j.compind.2018.01.014 | Computers in Industry |
Keywords | Field | DocType |
Information systems,Approximations,Rough set,Divergence measure,Fuzzy rough set,Feature selection | Information system,Data mining,Data set,Vagueness,Feature selection,Fuzzy logic,Control engineering,Fuzzy set,Rough set,Engineering,Salient | Journal |
Volume | ISSN | Citations |
97 | 0166-3615 | 4 |
PageRank | References | Authors |
0.38 | 12 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
T. K. Sheeja | 1 | 4 | 0.38 |
A. Sunny Kuriakose | 2 | 4 | 0.38 |