Prediction of Churn Behaviour of Bank Customers Using Data Mining Tools

Abstract

The customer churn is a common measure of lost customers. By minimizing customer churn, a company can maximize its profits. Companies have recognized that existing customers are the most valuable assets. Customer retention is critical for a good marketing and a customer relationship management strategy. The prevention of customer churn through customer retention is a core issue of Customer Relationship Management (CRM). The paper presents churn prediction based on data mining tools in banking. In this paper, a study on modeling purchasing behavior of bank customers in Indian scenario has been attempted. A detailed scheme is worked out to convert raw customer data into meaningful and useful data that suits modeling buying behavior and in turn, converts this meaningful data into knowledge for which predictive data mining techniques are adopted. In this analysis, the researchers have experimented with 2 classification techniques namely CART, and C 5.0. The prediction success rate of Churn class by CART is quite high but C 5.0 had shown poor results in predicting churn customers. However, the prediction success rate of Active class by C 5.0 is more effective than the other technique. However, for reaping significant benefits, the models have predicted the churn behavior.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 30 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.