Movie Recommendation System Based On Emotions

In this paper, we offer an engaging emotion-based movie recommendation system. With our approach, people may effortlessly convey their feelings through natural language in a search box. Using cutting-edge natural language processing tools, we examine and understand the user’s emotional condition. Its recommendation structures make use of strategies. The first technique is content-based filtering, which generates suggestions based on a variety of criteria, including actors, directors, and movie-related content. The algorithm makes recommendations for films based on an analysis of these features and films that the user has expressed interest in. Using the TF-IDF Vectorizer from the sci-kit-learn module, we apply TF-IDF vectorization to the anticipated genre to improve the precision of genre-based movie recommendations. With the help of this vectorization technique, we depict the anticipated genre. We used cosine similarity to recommend 10 movies that match the user’s preferences.