Comparison of Sentiment Analysis Evaluation Regarding International Mobile Equipment Identity Blocking Between the K-Nearest Neighbor and Support Vector Machine Methods

The growth of social media users is accelerating rapidly, thus encouraging a variety of opinion sentiments regarding events and issues, as well as dilemmas on social media. Besides that, social media has become the primary medium for people to express their opinions, including on Twitter. By considering that opinion data on social media can be converted into valuable information using sentiment analysis, this research collects mining technology from Twitter media and estimates which sentiments are more dominantly tweeted on Twitter, whether positive or negative. The results of this research provide a solution. Apart from that, bear in mind that sentiment analysis has recently attracted the attention of researchers; therefore, this research aims to conduct a sentiment analysis of user opinions on Twitter regarding IMEI blocking by customs. The methods used for sentiment analysis in this research are the K-Nearest Neighbor (K-NN) and Support Vector Machine (SVM) classification methods. This research found that the K-NN method with a value of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$K=3$</tex> was accurate at 78 %, and the SVM method with a linear kernel was correct at 71 % in classifying whether positive or negative sentiment was more dominant in tweets on Twitter.