Application of Fuzzy Theory and Data Mining on Customer Relationship in Securities Industry

• Main Authors: Wen-Ling Hung, 洪汶鈴
• Other Authors: Chui-Yi Chiu
• Format: Others
• Language: zh-TW
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/69692151431498421435

碩士 === 國立臺北科技大學 === 生產系統工程與管理研究所 === 91 === The cost of obtaining a new customer is higher than holding an old customer. Since customers are their most valuable assets, enterprises have to create competitive advantages by enhancing customer relationship management. Securities industry has complete customer trading data and can apply data mining technology to analyze customer trading behavior. We can further divide customers into groups according to their trading data characteristics. Meanwhile, the maketing planning can be made properly. This study uses customer historical trading data of a brokerage company. The data are first merged, converted and edited to explore their fuzzy relation and fuzzy models. Fuzzy inference is utilized for customer segmentation. Analysis of variance are used to verify the significance among our proposed fuzzy method , SOM + K-means. The result shows that our fuzzy model is superior to SOM+K-mean