Framework for mining data of credit card user's personal attitude

• Main Authors: Hui-wen, Peng, 彭慧雯
• Other Authors: 吳忠敏
• Format: Others
• Language: zh-TW
• Published: 2001
• Online Access: http://ndltd.ncl.edu.tw/handle/61803949040352349854

碩士 === 國立臺北科技大學 === 生產系統工程與管理研究所 === 89 === In Taiwan, the amount of outstanding credit has increased rapidly in last 10 years and banks get a lot of benefit in credit card business. For these reason, there should have a complete marketing strategies in credit card management. Finding knowledge about each different type customers become more and more important to keep competition It’s convenience for customers to have several credit card, but for bank that take more risk. It’s very important in credit card scoring to qualify user. Better scoring decrease the risk in losing benefit. Mining the different characterization in each type user could help increase customers’ satisfaction. In order to prove our idea, we design an algorithm to solve this problem. The goal of this study is using mining algorithms like discriminant analysis and neural network, to get knowledge in credit card. The first step is classifying customer to three types with cost and using rate. Then building and comparing the result of discriminant analysis model and neural network model to get the critical factors in scoring. In our research, we find neural net work have better explain and predict abilities than discriminant analysis.