Movie recommendation system on content based filtering
Akhilesh Kumar Singh, Md Jishan Ahmad, Hemant Shashi, Prabhat Sangri · 2025
A content-based movie recommendation system is presented in this paper. That [ 5 ] suggest movies based on their features, like genre. A dataset of 5,000 movies from IMDB, and the system combines this metadata into tag and processed it using text processing system technique, and calculate the similarity using cosine similarities. The system was able to find top 5 similar movies based on their actors, directors, genre, characters, The Movie recommend systems are essential for improving user experiences in the digital age because they direct viewers to content that suits their tastes. This study offers a thorough examination of content-based movie recommendation systems, an approach to recommend systems that analyzes the inherent characteristics of films, including cast, genre, plot synopsis, other metadata, to generate recommendations. Content-based systems avoid common problems like the cold start problem for new users by generating personalized recommend based only on individual user prefer as opposed to user interactions, which are the basis of collaborative filtering. [ 4 ] TF-IDF (term frequency inverse document frequency), feature extraction using natural language processing (NLP), and similarity metrics like cosine similarity to match are among the key approaches utilized in content-based systems that are examined.