Abstract:In practice, the Rough set theory based reduct is hardly used as the attributes are always indeterministic and data are polluted by noise. So we propose a new Information Entropy Based Reduct searching algorithm. The information entropy is introduced into the Reduct searching algorithm, so the statistical causalities between the attributes could be found and the Reduct's unavoidable sensitivity to the noise when the γ criterion is applied could be removed. Theoretical analysis and illustrated example showed that the rule based model resulting from this algorithm uses less attributes than those of model resulting from algorithms based on Rough sets γ criterion. At the same time, when the rules are applied, they get larger coverage, especially when noise and statistical causalities exist. In industrial applications, the data are always polluted by the noise and the causalities between the attributes are indeterministic, so this algorithm is more applicable than those based on Rough sets γ criterion.

Matrix Computation for Assignment Reduction and Rule Extraction in Incomplete Information Systems

Matrix Computation for Rule Extraction in Inconsistent Decision Tables

Matrix Computation for Data Cleaning and Rule Extraction in Knowledge System

Processing Methods for Incomplete Information Systems Based on Rough Sets

Rough computation based on similarity matrix

Upper/Lower Approximation Reduction-Based On Tolerance Relation

Ordered Assignment Reduction in Inconsistent Decision Tables

Rough Rule Extracting from Various Conditions: Incremental and Approximate Approaches for Inconsistent Data.

Rough Computational Methods Based on Tolerance Matrix

Constructing Rules Based on Rough Sets Theory and Information Entropy

Negative Rules and Reduction in Rough Sets Based on Similarity Relation

A parallel approximate rule extracting algorithm based on the improved discernibility matrix

A dynamic incremental rule extracting algorithm based on the improved discernibility matrix

A Heuristic Reduction Algorithm in Iis Based on Binary Matrix

A Constructing Method of Decision Tree and Classification Rule Extraction for Incomplete Information System

Credible Decision Rules Acquisition and E-relative Reduct in Incomplete Information Systems

On Knowledge Reduction in Inconsistent Decision Information Systems

Research On Method Of Fuzzy Sets Rules Reduction For Real Value Information System

bu wan bei xin xi xi tong zhong yi ge bian jing du cu cao ji mo xing

Knowledge Reduction In Incomplete Systems Based On Gamma-Tolerance Relation