PREDICTING CUSTOMERS LOYALTY USING RANDOM FOREST

• Main Author: Belva, Ahmad
• Format: Final Project
• Language: Indonesia
• Subjects: Ilmu alam dan matematika
• Online Access: https://digilib.itb.ac.id/gdl/view/52956
• Institution: Institut Teknologi Bandung

The research in this paper is about a Machine Learning algorithm namely Random Forest and its application in solving problems related to classification. The classification problem discussed in this paper is about predicting banks’ customers loyalty using parameters contained in the customers dataset. One of the prominent problems faced by banks in banking industry is about optimizing marketing cost during funding process. On banking sector, marketing cost is used to provide incentive and benefits for customers that already utilize said bank’s service and to attract new potential customers. By predicting customers’ loyalty, banks can optimize their marketing cost by adjusting incentive to existing customers proportional to their predicted loyalty. Random Forest algorithm will be used to build a model that could predict customers’ loyalty based on parameters provided in customers dataset. From experiments done in this research, the model obtained could predict customers’ loyalty sufficiently well represented by the obtained 96% Accuracy, 94% Precision, and 98% Recall.