Classification of Telemarketing Data using Different Classifier Algorithms

Abstract

Globalisation, growth of new technology usage and tough competition, made the banks to adopt new approaches to get competitive advantage to enlarge customer databases and also to generate customer satisfaction. In the present days the banks are trying to enhance customer base to meet their business targets for which they follow various approaches like Internet banking, Direct Tele Marketing, Mobile Banking, etc. Apart from banking services to customers, Banks are also selling Insurance policies to the customers through Tele Marketing and by which their business is expanding exponentially. In this paper, various Machine Learning Algorithms like Random Forest, Random Tree, Rep Tree, Naïve Baye’s, J48 Decision Tree before and after refinement of data and advanced Statistical techniques were applied for effective analysis of Bank’s Tele Marketing Data in order to enhance number of subscribing customer’s

Meta-Algorithms untuk Meningkatkan Kinerja Klasifikasi dalam Keberhasilan Telemarketing Perbankan

“…Hasil yang diperoleh Moro dkk adalah bahwa NN merepresentasikan hasil klasifikasi terbaik dengan pengukuran AUC = 0,8 dan ALIFT = 0,7. VenkateshYadav dkk [7] menggunakan model Decision Tree, Random Forest, NaïveBaye's dan Random Tree dengan hasil bahwa Random Tree memiliki performa terbaik dengan nilai root mean squared error (RMSE) dengan nilai 0. Elzhan dkk [4] juga membandingkan beberapa model klasifikasi untuk memprediksi telemarketing perbankan dengan hasil Random Forest memiliki nilai akurasi terbaik (90,9) daripada model KNN, LR, NN, NaïveBaye's dan SVM.…”