LRA: an accelerated rough set framework based on local redundancy of attribute for feature selection

TL;DR

This paper introduces the LRA framework, which accelerates rough set algorithms for feature selection by leveraging local attribute redundancy, maintaining accuracy while improving efficiency.

Contribution

The paper presents a novel, general-purpose LRA framework based on a new theorem about attribute stability, significantly speeding up rough set methods without accuracy loss.

Findings

LRA framework improves computational efficiency of rough set algorithms

Theoretical analysis confirms high efficiency and stability

No decrease in classification accuracy with LRA application

Abstract

In this paper, we propose and prove the theorem regarding the stability of attributes in a decision system. Based on the theorem, we propose the LRA framework for accelerating rough set algorithms. It is a general-purpose framework which can be applied to almost all rough set methods significantly . Theoretical analysis guarantees high efficiency. Note that the enhancement of efficiency will not lead to any decrease of the classification accuracy. Besides, we provide a simpler prove for the positive approximation acceleration framework.