Analyzing TikTok Account Reputation from Google Play Store Reviews using Sentiment Analysis and Machine Learning Approach

Variations in TikTok account reputation can affect user experience and platform image, so this research aims to address this issue. By implementing the development of a system to identify review sentiment towards TikTok accounts and provide reputation ratings. The method used is to collect 90,000 reviews from Google Play Store, text preprocessing to clean and prepare the data, and compare Machine Learning algorithms, including Naive Bayes, Support Vector Machine (SVM), and Neural Network to perform sentiment analysis. The result of this research is a model that can predict positive, negative, or neutral sentiments from user reviews of TikTok accounts, with accuracy results reaching 88% for Naive Bayes, 90% for SVM, and 88% for Neural Network. This shows that Sentiment Analysis and Machine Learning approaches can be effectively used to analyze TikTok account reputation from Google Play Store reviews. Factors that affect TikTok account reputation in this study include app performance, update quality, excess data consumption, content quality, and user interaction. Nonetheless, there is still room for improvement in sentiment prediction accuracy. Suggestions for future research are to include additional variables, such as user interaction or number of followers, as well as conduct experiments with more sophisticated algorithms to improve model performance.