Hits:

Indexed by:期刊论文

Date of Publication:2013-01-01

Journal:ICIC Express Letters, Part B: Applications

Included Journals:EI、Scopus

Volume:4

Issue:4

Page Number:913-920

ISSN No.:21852766

Abstract:This paper discusses and analyzes the measure of Support-Confidence as well as its limitation, and proposes a novel interestingness measure to improve the quality of the rules discovered from association rule mining. The proposed measure employs both objective measure and subjective measure to facilitate pruning rules. Besides, this interestingness measure incorporates context information into the process of rules filtering. To validate the feasibility of our measure, experiments are carried out on three different datasets. Experimental results show that the improved interestingness measure significantly reduces the number of redundant rules. Thus, the proposed measure outperforms the framework of Support-Confidence in term of effectiveness. And it has the advantages of reducing the creation of redundant rules and low-relation rules under the mobile ecommerce environment where contexts are critical to the success of the application. ? 2013 ISSN 2185-2766.