Home // INTELLI 2013, The Second International Conference on Intelligent Systems and Applications // View article

Apriori-with-Constraint for Flexible Association Rule Discovery

Authors:
Kittisak Kerdprasop
Phaichayon Kongchai
Nittaya Kerdprasop

Keywords: association rules; frequent itemset mining; data mining; association analysis;constraint logic programming.

Abstract:
Association rule discovery, or association mining, is one of the major data mining tasks that has gained much interest from researchers and general users. The knowledge obtained from association mining can be used to benefit business in many aspects such as recommend new products, design catalogs, manage sales promotion, and so on. But data processing for association rule discovery has expensive computing time because the relationships induced from data can be tremendously many more than those induced from other data mining tasks such as classification. As a consequence, most association mining software generally create so many rules from the association mining process and some of these rules are not beneficial to any users. To solve this useless rule mining problem, we propose to incorporate Apriori algorithm with constraint function for users to specify subset of association rules containing only interesting items. Besides specific items, users can also identify length of the association rules. Our two Apriori-with-constraint algorithms, called Association rule discovery with Constraints In Frequent itemset mining (ACIF) and Association rule discovery with Constraints After Frequent itemset mining (ACAF), are experimentally proven to be able to reduce processing time and also pruning a great number of useless rules.

Pages: 36 to 41

Copyright: Copyright (c) IARIA, 2013

Publication date: April 21, 2013

Published in: conference

ISSN: 2308-4065

ISBN: 978-1-61208-269-1

Location: Venice, Italy

Dates: from April 21, 2013 to April 26, 2013