Sentiment Analysis of Mobile Legends: Bang Bang User Reviews Using Machine Learning Models
Nafiatun Sholihah, Bima Pramudya Asaddulloh, Afrig Aminuddin, Jeeva Ekanayake, Ferian Fauzi Abdulloh, Majid Rahardi · 2025
This paper presents a sentiment analysis of user reviews for the “Mobile Legends: Bang Bang” application using data scraped from the PlayStore platform. The dataset comprises 10,000 user comments and ratings ranging from 1 to 5. Several machine learning models were evaluated, including Gaussian Naive Bayes, Multinomial Naive Bayes, Bernoulli Naive Bayes, K-Nearest Neighbors (KNN), and Random Forest, using accuracy, precision, recall, F1-score, and confusion matrices as evaluation metrics. The results indicate that Bernoulli Naive Bayes outperformed the other models, achieving an accuracy of 0.825, precision of 0.819, recall of 0.825, and an F1-score of 0.817. The model showed a strong balance in classifying positive and negative reviews, making it the most effective choice for sentiment analysis. Comparisons with KNN and Random Forest revealed that Bernoulli Naive Bayes performed superiorly in sentiment classification tasks.