Analisa Sentimen terhadap Twitter Pemilu 2024 menggunakan Perbandingan Algoritma Naïve Baiyes

Abstrak

In the digital era, sentiment analysis has become essential for understanding public opinion on various issues, including general elections. In the context of the 2024 General Election (Pemilu), this study aims to analyze sentiments expressed on the Twitter platform regarding the event. A primary classification algorithm, Naïve Bayes, was used to classify sentiments into positive, negative, and neutral categories and compare its performance. Twitter data was collected using a crawling technique during the 2024 election campaign period and used as the dataset. The data was then processed to remove noise and underwent text preprocessing, including tokenization, stemming, and stop word removal. Subsequently, the Naïve Bayes algorithm was applied to classify the sentiment of the collected tweets. Naïve Bayes, with its probabilistic approach and feature independence assumption, offers a fast and straightforward solution for classification tasks. The analysis results show that the algorithm was able to classify sentiments effectively. In tests using a separate test set, Naïve Bayes achieved an accuracy of approximately 82%. However, this algorithm has strengths and weaknesses that must be considered in the context of sentiment analysis on Twitter related to the 2024 election. For example, Naïve Bayes is more efficient in terms of time and resources. The study concludes that although Naïve Bayes produced accurate results, selecting the best algorithm depends on specific analysis needs, such as processing speed and resource availability. Further research is recommended to explore hybrid methods and deep learning techniques to enhance the accuracy and efficiency of sentiment analysis on social media platforms. The processed data consisted of 1,500 tweets. This study shows that the classification of Twitter data using the Naïve Bayes algorithm achieved an accuracy of 80%.