Sentiment Analysis of Online Game Clash of Clans Reviews Using the K-Nearest Neighbor Method

Abstract

Clash of Clans is a popular strategy game with millions of players worldwide. User reviews for this game are available on various online platforms. Sentiment analysis of these reviews can provide valuable insights into players' experiences and opinions. In this study, the researchers used the K-Nearest Neighbor (KNN) algorithm to classify the sentiment of Clash of Clans player reviews collected from the Google Play Store. Experimental results show that with a 60:40 training and testing data split, the KNN model was able to classify review sentiment with an accuracy of 64.52%, a precision value of 68.4%, a recall value of 88%, and an F1-score of 76.97%. The application of TF-IDF word weighting produced high accuracy at k-2 with an accuracy of 95.55%, precision of 96.16%, recall of 95.55%, and F1-score of 95.59%. These results indicate that KNN can be an efficient tool for analyzing player sentiment towards the Clash of Clans game.