Studi Sentimen Komunitas eSport Mobile Legends di Media Sosial X Menggunakan NLP

Abstract

Mobile Legends is one of the most popular eSports games in Indonesia, with an active fan community across various social media platforms. This study aims to analyze fan sentiment towards Mobile Legends regarding tournament-related topics, heroes, and feature updates by utilizing big data from Twitter (now known as X). Data was collected using a crawling technique on Twitter and then analyzed using Natural Language Processing (NLP) to identify positive, negative, and neutral sentiments in the gathered tweets. Out of 246 tweets, negative sentiment dominated (55%), followed by positive (39%) and neutral (7%). Criticism was mainly related to lag, bugs, and features, although some players expressed positive appreciation. Statistical analysis showed a Pearson correlation value of -0.975, indicating a very strong negative relationship between sentiment categories and the number of tweets. However, this relationship was not statistically significant (P-value 0.141). These findings provide valuable insights for developers in improving the quality of updates. Further studies with larger datasets and additional variables are recommended.