Sentiment Analysis with LSTM Recurrent Neural Network Approach for Movie Reviews using Deep Learning

In this research work, an integrated sentiment analysis system is designed specifically for movie reviews. Leveraging LSTM-based sentiment analysis, Proposed implementation simplifies model training using pre-trained LSTM models, resulting in robust sentiment prediction capabilities. The architecture emphasizes abstraction, separating frontend and backend components, optimizing data flow with JSON formatting and Pickle Files, and providing a user-friendly interface. By highlighting the role of abstraction in enhancing user experiences and system efficiency, the research reveals the successful synergy of DL techniques, architectural design, and abstraction principles. The LSTM model achieved testing accuracy of 87.53%.