CUSTOMER SEGMENTATION BASED ON CUSTOMER LIFETIME VALUE USING DATA MINING TECHNIQUE

• Main Author: FIKRI
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/41948
• Institution: Institut Teknologi Bandung

value over a period of time. It can be referred to as the present value of the income that will be generated by individuals. CLV determines customer value for the company during the customer's life cycle. CLV is used as the basis for customer segmentation in an effort to obtain customers that are profitable for the company. Determination of CLV value in this study uses the Recency, Frequency and Monetary (RFM) model which has been widely applied in various fields, especially in the field of marketing. The RFM model uses variables about the most recent purchase time (Recency), how many times the customer makes a purchase (Frequency), and the average money spent (Monetary). This study uses the RFM model as a basis for segmenting and determining CLV values. The RFM model is integrated with data mining techniques to support customer segmentation and classification. Data mining techniques used are Self Organizing Maps (SOM), K-Means, and Decision Tree C4.5. Research uses activity data and profiles of customers who have transacted shopping online. The SOM algorithm is used as a determinant of the number of clusters. After the number of clusters is formed, segmentation is done using the K-Means algorithm. Furthermore, classification is done using the Decision Tree C4.5 algorithm. Based on the results of data processing as many as 558 respondents, obtained five consumer segments, namely New Customer, Superstar, Typical Customer, Dormant Customer, and Occational Customer. Three classification categories, namely High Value Customer, Medium Value Customer, and Customer at Risk. The measure of goodness of the segment formed is shown by the Davies-Bouldin index of 0.472 and the classification accuracy is 70.9%.