Improving the Performance of Association Classifiers by Class Prioritization

機構典藏 > College of Engineering > Graduate Institute & Department of Computer Science and Information Engineering > Journal Article > Item 987654321/89148

Please use this identifier to cite or link to this item: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/89148

Title:	Improving the Performance of Association Classifiers by Class Prioritization
Authors:	Chen, Chunhao;Chiang, Dingan;Chen, Chihyang
Contributors:	淡江大學資訊工程學系
Keywords:	Associative Classification Algorithms;Association Rule;Ranking;Rule Dependency
Date:	2012-02-01
Issue Date:	2013-05-22 11:12:34 (UTC+8)
Publisher:	Bethel: Binary Information Press
Abstract:	Although different associative classification algorithms have been proposed, as pointed out by [8], none of the available associative algorithm considers the rule dependence problem that will directly influence the classification accuracy of the associative classifier. Since finding the optimal execution order of class association rules (CARs) is a combinational problem, in this paper, instead of finding the optimal execution order of CARs, we propose a polynomial time algorithm to re-rank the execution order of CARs by class priority to reduce the influence of the rule dependency problem. Consequently, the performance (the classification accuracy and recall rate) of the associative classification algorithm can be improved. The experimental results show that the association classifier with our method can get better classification result than that of the association classifier without considering the rule dependence problem.
Relation:	Journal of Computational Information Systems 8(4), pp.1697-1712
Appears in Collections:	[Graduate Institute & Department of Computer Science and Information Engineering] Journal Article

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	245	View/Open