Mining potentially more interesting association rules with fuzzy interest measure |
| |
Authors: | Wei-Min Ma Ke Wang Zhu-Ping Liu |
| |
Affiliation: | (1) School of Economics and Management, Tongji University, Shanghai, 200092, China;(2) School of Economics and Management, Beijing University of Aeronautics and Astronautics, Beijing, 100083, China |
| |
Abstract: | The association rules, discovered by traditional support–confidence based algorithms, provide us with concise statements of
potentially useful information hidden in databases. However, only considering the constraints of minimum support and minimum
confidence is far from satisfying in many cases. In this paper, we propose a fuzzy method to formulate how interesting an
association rule may be. It is indicated by the membership values belonging to two fuzzy sets (i.e., the stronger rule set
and the weaker rule set), and thus provides much more flexibility than traditional methods to discover some potentially more
interesting association rules. Furthermore, revised algorithms based on Apriori algorithm and matrix structure are designed
under this framework. |
| |
Keywords: | |
本文献已被 SpringerLink 等数据库收录! |
|