Comparison of potential telemarketing customers predictions with a data mining approach using the MLPNN and RBFNN methods

Abstract

Bank as a company that has many customers who conduct transactions every day, of course, has data that is increased continuously. In this paper, customer transaction data has been predicted to find potential customers in deposit offer. Data mining approach has been performed to classify potential customers for marketing through telemarketing. This data analysis can be used as a consideration in determining marketing strategy decisions for marketing managers. 15,713 data with 13 class attributes and 1 target class were obtained from UCI Machine Learning repository. The data was divided into 70% training data and 30% test data. Feedforward method of Artificial Neural Network was used to classify customer data. Multilayer Perceptron Neural Network and Radial Basis Function Neural Network were used to obtain optimal classification results. The result of this classification predicted potential customers to subscribe deposits. The result of this study indicated that the Radial Basis Function Neural Network method with 95.3% accuracy and 96.4% sensitivity was a better method compared to the Multilayer Perceptron Neural Network method with 88.0% accuracy and 99.4% sensitivity.

Access to Document

Other files and links

• Link to publication in Scopus

Fingerprint