Rearrange the Rules of Associative Classification using Simulated Annealing and Genetic Algorithm

Abstract:

Associative Classification (AC) is a new approach in data mining; it takes great attention against traditional classification methods like C4.5, because building the classifier using AC approach is more accurate and easy to interpret for end-user. In the recent years a lot of algorithms based on AC were proposed to enhance the accuracy of the classifier. In this paper, we proposed a new algorithm that utilizes the Simulated Annealing (SA) and Genetic Algorithm (GA) to build more accurate classifier. The proposed method rebuilds the classifier more than one time using SA and GA to reach the better classifier. The proposed algorithm tested on seven datasets from UCI Machine Learning Repository, the experimental results show that our algorithm gives better accuracy than CBA in most data sets.