JP Journal of Heat and Mass Transfer
Special Volume, Issue I, Advances in Mechanical System and ICT-convergence, Pages 55 - 69
(June 2018) http://dx.doi.org/10.17654/HMSI118055 |
|
REDUNDANT ASSOCIATION RULE MINING REDUCTION FOR ROSEWOOD CRIME ARREST PLANNING
Wararat Songpan, Ngamnij Arch-in and Rit Loaphanom
|
Abstract: Nowadays most applications for planning have applied the association rule mining algorithm. However, the disadvantage of this algorithm encountered the redundant patterns of the rules. For the implementation of the fundamental association rule mining algorithm (Apriori and FP-Growth), it was found the problem that there were redundant rules which did not exactly serve the user’s requirement for the real planning. The operating process was evaluated in terms of the support, the confidence, and the lift. Thus, this study proposed the redundant association rule mining reduction so that the user could apply it for planning more efficiently and instantly. Additionally, this algorithm has been applied to the real case study of the rosewood crime arrest planning, which was the wood with the legal restriction. The results of the study revealed that it could practically reduce the redundant rules and outputs of the rule patterns as the user’s requirement by 75%. The accuracy and precision of the outputs were increasingly higher by 95% as compared to the fundamental algorithm which output the redundant association rules to the user for planning. |
Keywords and phrases: association rule mining, redundant rule mining, crime arrest planning. |
|
Number of Downloads: 289 | Number of Views: 2718 |
|