Online movie streaming platforms have changed the current pattern of watching movies. Besides providing convenience in watching anywhere and anytime, this service is provided at a lower cost to moviegoers. The increase in moviegoers on online streaming platforms has resulted in easy-to-find reviews. This review can determine whether they decide to watch the film or not. The moviegoers' reviews can be easily and quickly found for analysis using sentiment analysis to find a film's worthiness. This study used sentiment analysis in classifying Twitter data predictions using the Multinomial Naive Bayes (MNB) and Support Vector Machine (SVM). In the sentiment analysis of labeling with positive and negative categories, a distilled version of BERT (DistilBERT) was used in this study. With a little human assistance in preprocessing, the model worked objectively with an overall accuracy performance on the confusion matrix of 64.50% for the Multinomial Naive Bayes model and 64.12% for the Support Vector Machine model. Performance evaluation was also carried out by calculating the cross-validation accuracy, which resulted in an accuracy of 72.38% for the MNB. Meanwhile, the SVM model obtained an accuracy of 70.19%.