GoPay is an electronic wallet service that can be used to pay for a number of Gojek app services. As the number of users increases, reviews on the Google Play Store have emerged as an important data source regarding the level of satisfaction with the app. Therefore, based on existing review data, sentiment analysis can be used to find trends in user opinions. The purpose of this study is to compare the performance of the Support Vector Machine and Naïve Bayes algorithms in classifying the sentiment of GoPay app user reviews. This research method uses the Knowledge Discovery in Database (KDD) framework which has 6 stages in the process, namely data selection, preprocessing, transformation, data mining, evaluation, and knowledge presentation. All stages of analysis were carried out using Google Colab with the help of the python programming language. The results show that the SVM method has better performance than the Naïve Bayes method. SVM produces an accuracy value of 78.31%, precision of 78.48%, and recall of 78.31%, while Naïve Bayes produces an accuracy of 64.74%, precision of 63.79%, and recall of 95.80%. Although Naïve Bayes has a higher recall value, over all the SVM method shows more optimal performance in classifying the sentiment of GoPay app user reviews.