<![CDATA[Every app has strengths and weaknesses that can influence various responses from users, including levels of satisfaction and disappointment that are often expressed through reviews on the Google Play Store. On this platform, the ratings and reviews feature allows users to give their opinions and experiences on the apps they use. One example of an application that is popular among the public is WhatsApp. The purpose of this research is to measure users' opinions and views on the WhatsApp application using the K-Nearest Neighbor algorithm. The data used in this study includes 1000 data, with 669 positive opinions and 331 negative opinions on the application. The process of dividing training data and test data was carried out through several experiments with three different ratios, namely 70:30, 80:20, and 90:10. From the results of this test, the best model was obtained in the scenario of dividing training data and test data with a ratio of 90:10 resulting in accuracy reaching 84%, precision value of 87.65%, recall of 92.21%, and f1-score of 89.87% for the positive class. While in the negative class, the precision value reached 68.42%, recall reached 56.52%, and f1-score reached 61.90% at K = 14 and Threshold = 20.]]>
Copyrights © 2023
