Support Vector Machine Method for Sentiment Analysis of Threads Applications on the Google Play Store

Abstract

This research started with user interest in the latest Twitter app and the Google Play Store's Threads app.The primary goal was to apply the Support Vector Machine (SVM) technique to analyze user evaluations for sentiment, both positive and negative.The collected data went through a preprocessing process that included cleaning, casefolding, tokenizing, stop word removal, stemming, and filtering. After going through preprocessing data as many as 1000 comments were implemented into the Support Vector Machine method showing 54.1% positive sentiment and 45.9% negative sentiment. The accuracy value of 81.19% and the confusion matrix results from the values of the TP (True Positive), TN (True Negative), FP (False Positive), and FN (False Negative) variables are also displayed by the testing. Based on the accuracy value acquired, it can be inferred that the Threads application receives positive feedback and is likely to be improved to yield even better outcomes.