β-Interval attribute reduction in variable precision rough set model

Authors:
Jie Zhou;Duoqian Miao
Affiliations:
Tongji University, Department of Computer Science and Technology, 201804, Shanghai, People’s Republic of China;Tongji University, Department of Computer Science and Technology, 201804, Shanghai, People’s Republic of China
Venue:
Soft Computing - A Fusion of Foundations, Methodologies and Applications - Special issue on advances in computational intelligence and bioinformatics
Year:
2011

Citing 0
Cited 1

Quantitative information architecture, granular computing and rough set models in the double-quantitative approximation space of precision and grade

Information Sciences: an International Journal

Quantified Score

Hi-index	0.00

Visualization

Abstract

The differences of attribute reduction and attribute core between Pawlak’s rough set model (RSM) and variable precision rough set model (VPRSM) are analyzed in detail. According to the interval properties of precision parameter β with respect to the quality of classification, the definition of attribute reduction is extended from a specific β value to a specific β interval in order to overcome the limitations of traditional reduct definition in VPRSM. The concept of β-interval core is put forward which will enrich the methodology of VPRSM. With proposed ordered discernibility matrix and relevant interval characteristic sets, a heuristic algorithm can be constructed to get β-interval reducts. Furthermore, a novel method, with which the optimal interval of precision parameter can be determined objectively, is introduced based on shadowed sets and an evaluation function is also given for selecting final optimal β-interval reduct. All the proposed notions in this paper will promote the development of VPRSM both in theory and practice.