Authors: Zailani Abdullah; Tutut Herawan; Mustafa Mat Deris
Addresses: Computer Science Department, Universiti Malaysia Terengganu, 21030 Kuala Terengganu, Terengganu, Malaysia ' Faculty of Computer System and Software Engineering, Universiti Malaysia Pahang, 26300 Kuantan, Pahang, Malaysia ' Faculty of Computer Science and Information Technology, Universiti Tun Hussein Onn Malaysia, 86400 Batu Pahat, Johor, Malaysia
Abstract: Mining of least association rules from large databases has received a great attention in knowledge discovery. These rules are very useful especially in tracing the unexceptional events or situations. Until this moment, the ratio of studies in this area is still unbalanced as compared to mine frequent rules. The difficulties level of mining these rules as compared to frequent rules are different since it involves with the excessive in computational costs, rather complicated and entailed a dedicated measurement. Therefore, this paper proposed an efficient model called critical least association rule (CLAR) to mine the significant rules so called critical least association rules. Several experiments with real and UCI datasets has shown that the CLAR successfully in producing the critical least association rules, up to 1.5 times faster and less 96% complexity than benchmarked FP-growth algorithm.
Keywords: significant rules; critical least association rules; knowledge discovery; association rule mining; data mining.
International Journal of Innovative Computing and Applications, 2013 Vol.5 No.1, pp.3 - 17
Available online: 28 Feb 2013 *Full-text access for editors Access for subscribers Purchase this article Comment on this article