Neural Network-Based Sentiment Analysis and Emotional Data Mining of Movie Reviews

With the increasing emphasis on spiritual and cultural enrichment, the film industry has seen remarkable growth, leading to a heightened demand for high-quality cinematic productions. Movie reviews, as a vital channel for capturing audience perspectives, provide valuable insights into films' strengths and areas for improvement through detailed ratings and commentary. This thesis focuses on extracting reviews from the Internet Movie Database (IMDb), a prominent English-language film review platform, and organizing them systematically within a database. After comprehensive data cleaning, advanced analytical techniques, including word frequency analysis and neural network-based sentiment classification, will be employed to interpret the reviews. The results, such as word clouds, sentiment distributions, and emotional trends, will be visualized via an interactive front-end interface. This approach aims to offer a nuanced understanding of audience sentiment and language patterns, contributing to the development of intelligent tools for movie evaluation and cultural analysis.