An attribute reduction algorithm in a decision based on improved entropy

Nguyễn Long Giang, Vũ Đức Thi
Author affiliations

Authors

  • Nguyễn Long Giang
  • Vũ Đức Thi

DOI:

https://doi.org/10.15625/1813-9663/27/2/500

Abstract

In rough set theory, many attribute reduction algorithms based on information entropy have been proposed. Although this algorithms reduce the time complexity, they do not obtain the minimal reduction in inconsistent decision tables. In this paper, we propose an improved entropy and we prove that the attribute reduction based on this entropy is equivalent to Pawlak's reduction in inconsistent decision tables. As a result, a complete heuristic algorithm was designed to find the attribute reduction based on improved entropy and its time complexity is O(|C|2 |U|), where |C| is the number of condition attributes, and |U| is the number of objects.

Metrics

Metrics Loading ...

Published

26-04-2012

How to Cite

[1]
N. L. Giang and V. Đức Thi, “An attribute reduction algorithm in a decision based on improved entropy”, JCC, vol. 27, no. 2, pp. 166–175, Apr. 2012.

Issue

Section

Computer Science