Predicting Customer Loyalty On Fast Moving Consumer Goods Using C4.5 Classification Method

Abstract

Instant noodles are one of the food products from the Fast Moving Customer Goods (FMCG) industry which is a fairly large industry in Indonesia. However, competition is inevitable. So to win the competition from other companies, company management is required to determine a strategi to maintain customer loyalty. Therefore, the purpose of this study is to create an application to predict customer loyalty and determine the influential attributes by applying Data Mining Classification in the form of a desicion tree. The application method used in Classification for prediction is the C4.5 method. In the C4.5 algorithm, entropy and information gain are calculated where customer loyalty is the attribute of destination (class), while price, packaging, taste, cariety, advertising, distribution, and quality are the source attributes to obtain the root node and other nodes. The results of the study show that the application using the C4.5 method produces an accuracy of 95.5%, so the C4.5 method can be used to assist the management of instant noodle companies in order to determine strategies to maintain consumer loyalty.