<![CDATA[Abstract—During the COVID-19 pandemic that interferesnormal life around the world, people have an obligation to stay athome and quarantine themselves. This has led to an increase inthe consumption of entertainment, especially online gaming whichis known to be less harmful than other stress and aversiveemotions. And Genshin Impact is one of the online games that wonGoogle Play's the Best Game of 2020 award when pandemichappening. Released in September 2020 by China video gamedeveloper, miHoYo. Co., Ltd, Genshin Impact has been a hottrend on the microblogging platform, Twitter. The purpose of thisresearch is to provide information regarding people's opinionemotion in their tweets toward Genshin Impact and thisinformation will be a helpful resource for game improvement andcan be used as reference of future research. By using sentimentanalysis to help analyze the emotion contained in the text, theresult will be categorized into three categories: positive, negative,or neutral sentiment. The data is gained through text mining thenwill be processed as text classified using Naive Bayes algorithm.Thus, the model will be going through evaluation of model'sperformance to measure how accuracy it is. The result of it statedthat the best ratio between training and test set is 60:40 with71.80% test accuracy, yet the accuracy between 3 others ratio isnot much difference. That’s why using hyperparameter tuningcan find the optimal result. After finding the optimal result, thehighest result it can get is 72.14%. Besides that, people on Twittermostly perceive the game in neutral sentiment. Keywords—Sentiment Analysis, Genshin Impact, Naïve Bayes Classifier]]>
Copyrights © 2024
