ScholarLens Back to search
2007•Computer Engineering and Applications JournalRequires access

Decision table attribute reduction algorithm based on extended information entropy

Yunsong Zhao

Open publisher page 1 citations

Abstract

The rough set theory is discussed in the light of an extended information view.A core attribute computation algorithm of decision table based on extended information entropy is proposed.And a from-bottom-to-top decision table attribute reduction algorithm which takes the significance of the attribute as the heuristic information is designed.Meanwhile,the including value of attributes is adopted as the second standard to choose attribute in order to make the reduction faster.The new algorithm EIEAAR can deal with both the consistent and inconsistent decision tables,and integrate the core attribute computation and non-core attribute reduction in a whole.At last,the complexity of the algorithm is analyzed and two kinds of examples are taken to test the validity of the algorithm.The experiment shows that the algorithm is valid.

About this research paper

What this paper is about

The rough set theory is discussed in the light of an extended information view.A core attribute computation algorithm of decision table based on extended information entropy is proposed.And a from-bottom-to-top decision table attribute reduction algorithm which takes the significance of the attribute as the heuristic information is designed.Meanwhile,the including value of attributes is adopted as the second standard to choose attribute in order to make the reduction faster.The new algorithm EIEAAR can deal with both the consistent and inconsistent decision tables,and integrate the core attribute computation and non-core attribute reduction in a whole.At last,the complexity of the algorithm is analyzed and two kinds of examples are taken to test the validity of the algorithm.The experiment shows that the algorithm is valid.

Why it matters

OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

The rough set theory is discussed in the light of an extended information view.A core attribute computation algorithm of decision table based on extended information entropy is proposed.And a from-bottom-to-top decision table attribute reduction algorithm which takes the significance of the attribute as the heuristic information is designed.Meanwhile,the including value of attributes is adopted as the second standard to choose attribute in order to make the reduction faster.The new algorithm EIEAAR can deal with both the consistent and inconsistent decision tables,and integrate the core attribute computation and non-core attribute reduction in a whole.At last,the complexity of the algorithm is analyzed and two kinds of examples are taken to test the validity of the algorithm.The experiment shows that the algorithm is valid.

Key concepts: Rough set, Attribute domain, Decision table, Computer science, Algorithm, Computation, Entropy (arrow of time), Reduction (mathematics)

Related papers

Related paper 1Related paper 2Related paper 3
Back to paper searchBrowse research topicsOriginal source
Decision table attribute reduction algorithm based on extended information entropy — Research Paper | ScholarLens