A Multi-Criteria Movie Recommendation System based on User Preferences and Movie Features

Abstract

In this research, we develop a multi-criteria movie recommendation system that provides personalised recommendations by taking into consideration both user preferences and movie aspects. To get over each method's specific drawbacks, the suggested system takes a hybrid approach that combines collaborative filtering with content-based techniques. The system uses collaborative filtering to capture user preferences based on historical ratings, while content-based methods analyze movie features such as genre, director, actors, and keywords to enhance the recommendation process. Additionally, we integrate various external data sources like movie reviews, social media sentiment, and box office performance to enrich the movie feature set. The system employs a weighted aggregation method to combine these criteria and generate a comprehensive recommendation score. The effectiveness of the proposed system is evaluated utilizing standard metrics including recall, precision, and F1-score on a publicly available dataset. The results demonstrate that our multi-criteria recommendation system effectively captures user preferences and provides more accurate and diverse recommendations compared to traditional single-criterion approaches.