Using Data Mining Techniques for Detecting the Important Features of the Bank Direct Marketing Data

Authors

  • Tuba PARLAR
  • Songül Kakilli ACARAVCI Mustafa Kemal University, Hatay

Abstract

Collection of customer information is seen necessary for development of the marketing strategies. Developing technologies are used very effectively in bank marketing campaigns as in many field of life. Customer data is stored electronically and the size of this data is so immense that to analyse it manually with a team of human analysts is impossible. In this paper, data mining techniques are used to interpret and define the important features to increase the campaign's effectiveness, i.e. if the client subscribes the term deposit. The bank marketing dataset from the University of California at Irvine Machine Learning Repository has been used for the proposed paper. We consider two feature selection methods namely Information Gain and Chi-square methods to select the important features. The methods are compared using a supervised machine learning algorithm of Naive Bayes. The experimental results show that reduced set of features improves the classification performance.

Keywords: Bank marketing, feature selection, machine learning methods, data mining, chi-square, information gain.

JEL Classifications: C80, C50, Y10, M31

Downloads

Download data is not yet available.

Downloads

Published

2017-04-14

How to Cite

PARLAR, T., & ACARAVCI, S. K. (2017). Using Data Mining Techniques for Detecting the Important Features of the Bank Direct Marketing Data. International Journal of Economics and Financial Issues, 7(2), 692–696. Retrieved from https://www.econjournals.com/index.php/ijefi/article/view/4580

Issue

Section

Articles