Investigation in Customer Value Segmentation Quality under Different Preprocessing Types of RFM Attributes

Authors

  • Nesma mahmoud Taher Faculty of Informatics & Computer Science, Helwan University, Cairo, Egypt
  • Doaa Elzanfaly Faculty of Informatics & Computer Science, British University in Egypt, 11873, Cairo, ,Egypt.
  • Shaimaa Salama Faculty of Informatics & Computer Science, Helwan University, 11795,Cairo, Egypt.

DOI:

https://doi.org/10.3991/ijes.v4i4.6532

Abstract


Customer value segmentation helps retailers to understand different types of customers, develops long term relationship with them, and hence increases their value and loyalty. This study aims to evaluate the quality of customer value segmentation based on two methods of preprocessing the RFM attributes. K-means clustering algorithm is used for the customer value segmentation based on the scored RFM and the actual value of RFM. The quality of the clustering results is tested using the Sum of Squared Error (SSE). Results obtained show that using the actual value of RFM in customer segmentation reduces the clustering error (SSE) and enhances the accuracy of segmentation than using the scored RFM.

Downloads

Published

2016-12-30

How to Cite

Taher, N. mahmoud, Elzanfaly, D., & Salama, S. (2016). Investigation in Customer Value Segmentation Quality under Different Preprocessing Types of RFM Attributes. International Journal of Recent Contributions from Engineering, Science & IT (iJES), 4(4), pp. 5–10. https://doi.org/10.3991/ijes.v4i4.6532

Issue

Section

Papers