Various studies on consumer purchasing behaviors have been presented and used in real problems. Data mining techniques are expected to be a more effective tool for analyzing consumer behaviors. However, the data mining method has disadvantages as well as advantages. Therefore, it is important to select appropriate techniques to mine databases.
The objective of this paper is to improve conventional data mining analysis by applying several methods including fuzzy clustering, principal component analysis, and discriminate analysis. Many defects included in the conventional methods are improved in the paper. Moreover, in an experiment, association rule is employed to mine rules for trusted customers using sales data in a fiber industry.