ANALISIS SENTIMEN MASYARAKAT PENGGUNA TWITTER MENGENAI CAPRES 2024 MENGGUNAKAN METODE NAIVE BAYES CLASSIFIER

This research aims to conduct sentiment analysis among the Twitter user community regarding the 2024 presidential candidates in Indonesia. The method
used in this research is the Naive Bayes Classifier, an approach commonly used in sentiment analysis to classify text into positive and negative categories. The text
data collected relates to presidential candidates in 2024. The data was then cleaned and labeled based on the sentiment contained in each tweet. Presidential processing processes include case folding, tokenizing, filtering and stemming.
After that, features from the text are extracted using the TF-IDF method to build a
naive Bayes classifier model. The dataset is then dividedinto two parts: 70% for model training and 30% for testing. The NBC model was trained using training
data and evaluated using testing data to measure its performance in classifying tweet sentiment related to the 2024 presidential candidates. The evaluation results show the model's accuracy, precision, recall and F1-score values. This research provides valuable insight into the public's views and sentiments towards the 2024 presidential candidates based on an analysis of tweets posted on Twitter
Keywords: Sentiment Analysis, Naive Bayes Classifier, Twitter, 2024 Presidential Candidates , Pre-processing, TF-IDF