Klasifikasi Tingkat Kepuasan Pengguna Gopay Menggunakan Algoritma C4.5 dengan Metode End User Computing Satisfaction (Eucs)

GoPay stands as one of the most widely adopted digital wallet services in Indonesia, yet its expanding user base does not consistently guarantee uniform satisfaction due to recurring technical issues and transaction security concerns. This study aims to evaluate user satisfaction levels across five dimensions of End User Computing Satisfaction (EUCS) Content, Accuracy, Format, Ease of Use, and Timeliness, and construct a classification decision tree model using the C4.5 algorithm to categorize satisfaction into Satisfied and Not Satisfied classes. Primary data were gathered through a Likert-scale questionnaire from 109 active GoPay users, validated for construct validity and reliability using SPSS, and pre-processed in RapidMiner. Data splitting was implemented with an 80:20 ratio, forming 88 training instances and 21 testing instances. The information gain ratio evaluation identified Ease of Use as the top-ranked attribute with a value of 1.000, establishing it as the root node of the decision tree model and yielding five distinct classification rules. Model evaluation using a confusion matrix achieved an overall accuracy of 71,43%, a weighted mean precision of 66,67%, and a weighted mean recall of 83,33%. The findings demonstrate that the C4.5 algorithm is effective in classifying GoPay user satisfaction, with Ease of Use serving as the most decisive determinant.