Analisis Sentimen Penonton Terhadap Film Sore Istri dari Masa Depan pada Twitter Menggunakan Algoritma Support Vector Machine

Abstract

The development of social media has encouraged people to express their opinions about cinematic works openly, quickly, and in real time. Twitter, now known as X, has become one of the most widely used platforms for sharing brief film reviews, generating a large volume of opinion data that is difficult to analyze manually. This study aims to analyze audience sentiment toward the film Sore: Istri dari Masa Depan based on tweet data, apply the Support Vector Machine algorithm in the sentiment classification process, and evaluate the performance of the resulting model. The research data were collected through a tweet crawling process using keywords related to the film Sore: Istri dari Masa Depan from July 2025 to April 2026. A total of 7,467 tweets were analyzed through text preprocessing, sentiment labeling, feature weighting using Term Frequency-Inverse Document Frequency, and classification using Support Vector Machine. Model evaluation was conducted using a confusion matrix with accuracy, precision, recall, and F1-score metrics, using an 80:20 training and testing data split. The results showed that 6,822 tweets were classified as positive sentiment and 645 tweets as negative sentiment. The model achieved an accuracy of 90.75%, indicating that it was able to classify audience sentiment effectively. This study contributes by providing a computational approach to mapping public reception of Indonesian films through social media data, while also offering a basis for film industry practitioners to evaluate audience responses and develop data-driven promotional strategies.