Public Sentiment Analysis on Dirty Vote Movie on YouTube using Random Forest and Naïve Bayes

Abstract

In early 2024, the film Dirty Vote attracted public attention, sparking discussions on YouTube. Understanding public sentiment towards this film is important for evaluating the reception of the work and its impact on public opinion. This study analyses 4,551 YouTube comments using the Random Forest and Naïve Bayes algorithms. The data was collected using the Apify platform, which allows the extraction of comment data based on video links and the desired amount of data. The analysis results show that the film received more negative comments than positive, reflecting the public's reception of the socio-political issues raised in the film. This dominance of negative sentiment is important for understanding how the film's message is received, which could influence marketing strategies and the film's reception in the digital media industry. This study also compares the effectiveness of both algorithms in sentiment analysis, with Random Forest being more effective at identifying positive sentiment, while Naïve Bayes is more efficient, though less accurate at capturing positive sentiment. These findings provide insights for developers and analysts in selecting the appropriate algorithm for sentiment analysis applications on social media.