Customers Classification using Recency Frequency, Monetary value (RFM), and K-means clustering algorithm

Ing. Mark Azietaku

Master's thesis

Abstract:

Data mining has changed how people see data and how much knowledge and information it may provide for business decisions to enhance competitiveness and profitability. Companies use data mining techniques with Customer Relationship Management (CRM) to better understand their customers' requirements and preferences. This study suggested using K-means algorithms for consumer segmentation in a public institution by utilizing fuzzy analytical network process (FANP) and RFM extended to RFMTL (Recency, Frequency, Monetary, Time, and Location). The established data mining process divided the client base into four groups: best clients, prospective and new customers, high-risk consumers, and lost customers. Marketing techniques were implemented based on the dynamics and traits of the clients in the segments to develop formidable strategies the organizations can use to attract, retain, and grow customers. Lastly, an assessment was conducted on the established strategies to ascertain their potential, correctness, consistency, and suitability for the business.

Abstract:

Data mining has changed how people see data and how much knowledge and information it may provide for business decisions to enhance competitiveness and profitability. Companies use data mining techniques with Customer Relationship Management (CRM) to better understand their customers' requirements and preferences. This study suggested using K-means algorithms for consumer segmentation in a public institution by utilizing fuzzy analytical network process (FANP) and RFM extended to RFMTL (Recency, Frequency, Monetary, Time, and Location). The established data mining process divided the client base into four groups: best clients, prospective and new customers, high-risk consumers, and lost customers. Marketing techniques were implemented based on the dynamics and traits of the clients in the segments to develop formidable strategies the organizations can use to attract, retain, and grow customers. Lastly, an assessment was conducted on the established strategies to ascertain their potential, correctness, consistency, and suitability for the business.

Keywords

Customer Relationship Management (CRM), Data mining, Algorithm, K-means, Recency Frequency and Monetary value (RFM), Recency Frequency Monetary value Time and Location (RFMTL), Classification, Segmentation, Fuzzy Analytical Network Process (FANP), Elbow Method, Silhouette Method

Language used: English

Date on which the thesis was submitted / produced: 5. 1. 2024

Thesis defence

• Date of defence: 30. 1. 2024
• Supervisor: doc. Ahad Zareravasan, PhD
• Reader: Ing. Michal Jirásek, Ph.D.

Masaryk University

Faculty of Economics and Administration

Master programme / field:
Business Management / Business Management