The development of increasingly sophisticated technology has made many changes and conveniences for humans in all things, including in the financial sector, cryptocurrency has emerged as a form of innovation in the world of finance. A cryptocurrency exchange is an electronic means for carrying out crypto asset trading transactions by sellers or buyers via websites or applications on mobile devices. Currently, many cryptocurrency exchange applications have poor service, security is not guaranteed, withdrawals take quite a long time, expensive admin fees, and other problems, so many Indonesians use reviews on Google Play Store to see user reviews before deciding to use the cryptocurrency exchange application. Many Indonesians are looking for information on cryptocurrency exchange applications that provide the best service for carrying out cryptocurrency buying and selling transactions. One of the cryptocurrency exchange applications that offers this is the Tokocrypto application based on reviews on the Google Play Store. This research aims to understand sentiment towards user reviews on the Tokocrypto application using the Naïve Bayes algorithm method for data classification. The data obtained is review data from February 2024, totaling 2000 reviews from the Google Play Store using Google Collaboratory. This research stage consists of data scraping using web scraping techniques, data labeling, preprocessing, TF-IDF weighting, implementation of the Naïve Bayes algorithm, and evaluation. The clean data obtained was 1000 with a total of 396 positive sentiments and 604 negative sentiments. The results of the sentiment analysis research using the Naïve Bayes algorithm method resulted in an accuracy level of 74.22%, precision of 63.25%, and recall of 81.40%.