<![CDATA[The phenomenon of e-sports has experienced significant developments along with the rapid development of information and communication technology during the industrial revolution 4.0. This industry has achieved several lucrative achievements and has become a popular topic that reaps a variety of comments on social media, one of which is Twitter. Sentiment analysis is useful for determining the tendency of public opinion to be positive or negative, so that later it can become structured information. The data used is 200 twitter comments. The processes carried out are crawling, preprocessing including cleaning, normalization, case folding, stopword removal, tokenizing, and stemming, carrying out sentiment labeling and weight calculations using TF-IDF. The classification method uses K-Nearest Neighbor, validation tests use 10-fold cross validation and accuracy calculations use the confussion matrix formula. The rapid miner tool is used to process the stages automatically. The final result obtained a value of k = 6 as the highest value with an accuracy of 66.00%, precision with a value of 67.34%, recall 78.06% and AUC with a value of 0.762. While the results of the sentiment classification obtained positive sentiment superior to negative sentiment. Based on these results, the K-Nearest Neighbor method is considered to be able to classify well and it is known that Indonesian people currently respond to the e-sports industry tends to provide more positive stigma and a good response.]]>
