Expected Confidence Dependency: A Novel Rough Set-Based Approach to Feature Selection

TL;DR

This paper introduces Expected Confidence Dependency (ECD), a new soft computing measure for feature selection in rough set theory that improves upon traditional binary dependency measures by incorporating confidence levels.

Contribution

The paper presents ECD, a novel dependency measure that assigns confidence-based contributions to equivalence blocks, enhancing feature selection in rough set theory.

Findings

ECD is formally proven to be normalized and monotonic.

ECD maintains compatibility with classical dependency measures.

The approach is invariant under structural and label-preserving transformations.

Abstract

This paper proposes Expected Confidence Dependency (ECD), a novel, soft computing-oriented, accuracy driven dependency measure for feature selection within the rough set theory framework. Unlike traditional rough set dependency measures that rely on binary characterizations of conditional blocks, ECD assigns confidence-based contributions to individual equivalence blocks and aggregates them through a normalized expectation operator. We formally establish several desirable properties of ECD, including normalization, compatibility with classical dependency, monotonicity, and invariance under structural and label-preserving transformations.