Analisis Sentimen Masyarakat terhadap Isu Pemalsuan Ijazah Joko Widodo di Media Sosial X Menggunakan Metode Naive Bayes

Abstract

The alleged diploma forgery case involving Joko Widodo attracted great attention on social media, especially on the X platform. This research aims to analyze people's association with the issue by utilizing the Naive Bayes algorithm. The data used in this study consists of 1,000 tweets obtained through web crawling techniques based on certain keywords. The dataset is divided into two segments, 80% used for training data and 20% used for testing data. Of the 800 data used for training, 299 tweets (37.4%) showed positive sentiment, and 501 tweets (62.6%) showed negative sentiment. The remaining 200 tweets were used to test the model. The Naive Bayes model was then trained with the training data to identify existing sentiment patterns, and then tested with the test data to assess its classification accuracy. The findings of this study indicate that negative sentiment is superior to positive sentiment regarding the issue of Joko Widodo's diploma forgery. This reflects the public's tendency to respond to politically sensitive issues in the digital world. The Naive Bayes method proved to be quite powerful in classifying text-based public opinion with high accuracy, so it has the potential to become a tool in mapping public views on socio-political issues.