Sentiment Analysis of Movie Reviews Using TF-IDF

Sentiment analysis is a computational approach for extracting and interpreting human emotions and opinions from textual data. Also known as opinion mining, it plays a crucial role in understanding public perception across various domains, including social media, online reviews, and customer feedback. The growing volume of user-generated content on digital platforms has increased the demand for efficient sentiment classification techniques. This study focuses on analyzing sentiment in movie reviews using machine learning algorithms, including Multinomial Naïve Bayes, Support Vector Machine, Random Forest, K-Nearest Neighbour, and Decision Tree. The models are evaluated based on their classification accuracy, highlighting their effectiveness in differentiating between positive and negative sentiments. Experimental results indicate that the Support Vector Classifier achieves the highest accuracy of 87.05%, whereas the Decision Tree performs the worst at 70.55%. The findings emphasize the importance of selecting appropriate machine learning models for sentiment analysis tasks. Further optimizations, including feature engineering and hyperparameter tuning, can enhance model performance and improve sentiment classification accuracy in real-world applications.