Implementation of Data Mining Using the C4.5 Algorithm to Predict Customer Satisfaction

Abstract

The C4.5 algorithm is one of many data mining methods that can generate rules for analyzing customer satisfaction. This process produces approximately four rules with influential attributes, namely Security, Timeliness, and Customer Service. The rapid changes in information technology have had a major impact on many aspects of life, including the business sector. One of the main transformations is the shift from traditional sales systems to digital ones, known as e-commerce. Tight competition in sales and maintaining customer loyalty makes customer satisfaction a crucial aspect to maintain good customer relationships. This research is expected to improve customer satisfaction and address issues related to it. In this context, the application of data mining is considered an appropriate tool for processing data in research to predict customer satisfaction using the C4.5 algorithm, as well as to test and validate information regarding customer satisfaction. The results of the research that has been carried out show that comparative testing on the classification model with the C4.5 algorithm produces a high level of accuracy in predicting data, namely with an accuracy value of 98.13%. For the recall test value, a value of 100.00% was obtained, while the precision value was recorded at 91.00%, and the ROC/AUC value reached 0.695%.