Exploring WeTV Application with Naïve Bayes, Decision Tree, and Random Forest Classifiers for Sentiment Analysis

• Published in: 2024 International Visualization, Informatics and Technology Conference (IVIT) pp. 35 - 42
• Main Authors: Mareby, Yeremia Steven Putra, Desanti, Ririn Ikana
• Format: Conference Proceeding
• Language: English
• Published: IEEE 07.08.2024
• Subjects: Accuracy; Analytical models; Bayes methods; Data models; Decision trees; digital content; machine learning algorithm; Machine learning algorithms; Random forests; Reviews; Sentiment analysis; Testing; WeTV application
• DOI: 10.1109/IVIT62102.2024.10692731

In 2023, Indonesia has experienced a notable increase in internet usage, particularly in the realm of digital content payments. This rise has consequently led to a substantial growth in the use of streaming applications, which are vital for providing entertainment to various communities by enabling content streaming on their devices. This research focuses on analyzing the general sentiments expressed in user reviews of WeTV, a popular streaming platform in Indonesia. To achieve this, the study employs three machine learning algorithms: Naïve Bayes, Decision Tree, and Random Forest. Among these algorithms, Naïve Bayes demonstrates the highest performance, achieving an accuracy of 79.1%, followed by Random Forest and Decision Tree. The results indicate that a significant portion of user sentiments towards the WeTV application are predominantly negative. The high accuracy of the Naïve Bayes model will be utilized for deployment on a basic website accessible to users, aiming to provide insights into user opinions and improve the overall user experience on the platform.