Data mining in the application of integrated marketing personal loan from a commercial bank in Taipei

• Main Authors: Bo-Qun, Chen, 陳博群
• Other Authors: Dr. Tian-Shyug, Lee
• Format: Others
• Language: zh-TW
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/3g2fs3

碩士 === 輔仁大學 === 金融與國際企業學系金融碩士在職專班 === 103 === 　　Financial Holding Company Act and financial liberalization cause the banking industry flourishing in Taiwan. Banks want to earn more market share and improve their profits. As personal loan is a simple financial product with high interest rate and profit margin, all banks cast lots of efforts trying to promote and gain market share from this product. 　　This research adopts data mining tools such as logistic regression, support vector machine and backpropagation network to build classification model for the personal loan credit scoring tasks. The proposed scheme will not only be effective to filter target customers, but also reducing the work loads of credit scoring department and improving the lending process. 　　According to the empirical study results, logistic regression has highest classification accuracy which outperforms those of backpropagation neural networks and support vector machines. The obtained results can provide useful references for financial institutions in designing better credit scoring strategies.