Predictive Analysis of Bank Telemarketing Results

Bo Chao ； Yuntong Lu ； Ziyi Wang

《Frontiers in Economics and Management》 1卷12期 (2020/12) Pp. 102-110

https://doi.org/10.6981/FEM.202012_1(12).0016

摘要

According to the data of 41,888 telemarketing time deposits of foreign banks, 17 variables related to bank customers, products and socio-economic attributes were used for model construction, and three models, CART, AdaBoost and SVM, were established using decision trees, Boosting and other methods. aspects are studied. By comparing the kappa values, sensitivity-specific curves, and ROC curves of each model, the Ada boost algorithm is found to be optimal, and suggestions are given to banks to improve their marketing success rate.

關鍵字

Bank Deposits ； Telemarketing ； Classification ； ROC ； AdaBoost Algorithm

參考文獻

Felicitas J. Detmer, Juan Cebral, Martin Slawski. A note on coding and standardization of categorical variables in (sparse) group lasso regression. 2020, 206:1-11.

P. H. Chen, C. J. Lin, and B. Schölkopf, A tutorial on v-support vector machines, Appl. Stoch. Models. Bus. Ind., 2005 21(2) pp. 111-136.

David L Olson, Dursun Delen, and Yanyan Meng, Comparative analysis of data mining methods for bankruptcy prediction.Decis, Support Syst., 2012 52(2) pp. 464-473.

Rongfang Gao, Gao Rongfang, Liu Zhanyu. An Improved AdaBoost Algorithm for Hyperparameter Optimization. 2020, 1631(1):012048-.

"Frontier Computing", Springer Science and Business Media LLC, 2020