Modern Research in Decision Making

Dynamic Customer Churn Modelling In Banking Industry Using Sequence Mining

Document Type : Original Article

Authors

Alireza Shebro 1
Elham Akhondzadeh Noughabi 2

1 1. MSc. in Faculty of Industrial and Systems Engineering, Tarbiat Modares University, Tehran, Iran

2 Assistant Professor, Department of Information Technology Engineering, Faculty of Industrial and Systems Engineering, Tarbiat Modares University, Tehran, Iran

Abstract
Accurately understanding customer behavior and predicting future trends are key priorities for organizations, particularly in the banking industry, where retaining existing customers is significantly more cost-effective than acquiring new ones. One of the major challenges in this context is the timely identification and prevention of customer churn, especially when dissatisfaction with services or gradual behavioral changes lead to disengagement. This study focuses on modeling dynamic customer churn by analyzing three years of transaction and account balance data from one of the top five banks in Iran. The proposed method employs sequence pattern mining to identify behavioral sequences that lead to churn. The extracted patterns are categorized into two main groups: dominant patterns that indicate high-risk churn situations or sustained loyalty, and high-confidence, low-support patterns that serve as effective tools for monitoring customers on the verge of churn. A key advantage of this method is its ability to generate interpretable if-then rules, which are easily understood and applied in real-world business environments. This analytical framework offers valuable insights for designing targeted interventions in customer relationship management.

Keywords

Customer Dynamics
Data Mining
Dynamic Customer Churn Modelling
Sequence Mining

Home

Guide for Authors

Submit Manuscript

Reviewers

Contact Us