Abstract

The rapid growth of social media users has led to the emergence of various sentiments and opinions related to events, issues, and dilemmas that develop on the platform. In addition, social media has become the primary medium for people to express their opinions, one of which is through Twitter. Given that opinion data on social media can be processed into valuable information through sentiment analysis and data mining methods, this study mines Twitter data to identify and estimate the most dominant sentiments in tweets, including positive, negative, and neutral sentiments. The results of this study provide a solution for analyzing the sentiment of user opinions on Twitter related to the People's Housing Savings program. It is worth noting that sentiment analysis has recently garnered increasing attention from researchers. Therefore, this study aims to conduct sentiment analysis using the classification methods of Support Vector Machine (SVM), Random Forest (RF), eXtreme Gradient Boosting (XGBoost), and Naïve Bayes (NB). The study findings reveal that the SVM method achieves the highest accuracy of 84%, followed by RF with 81% accuracy, XGBoost with 78% accuracy, and NB with 75% accuracy in classifying positive and negative sentiments, as well as neutral tweets on Twitter. Likewise, the precision, recall, and F1-score of SVM outperform RF and XGBoost.

Concepts :

Access to Document

SDGs

Citations by Year