Abstract—Association rule are important to retailers as a
source of knowledge to manage shelf, to plan an effective
promotion, and so on. However, when we are mining with
association rule discovery technique, we normally obtain a large
number of rules. To select only good rule is difficult. Therefore,
in this paper we propose the fuzzy search technique to discover
interesting association rule. The comparative result of fuzzy
versus non-fuzzy searches are presented in the experiment
section. We found that fuzzy search is more flexible than the
non-fuzzy one in finding highly constrained rules.
Index Terms—Fuzzy set, fuzzy search, membership function,
association rule mining.
The authors are with the School of Computer Engineering, Suranaree
University of Technology, Nakhon Ratchasima 30000, Thailand (e-mail:
Zaguraba_ii@hotmail.com).
[PDF]
Cite:Phaichayon Kongchai, Nittaya Kerdprasop, and Kittisak Kerdprasop, "The Fuzzy Search for Association Rules with Interestingness Measure," International Journal of Computer Theory and Engineering vol. 6, no. 6, pp. 490-494, 2014.
