Analyzing the Effectiveness of Collaborative Filtering and Content-Based Filtering Methods in Anime Recommendation Systems

Abstract

In the current digital era where content consumption via streaming platforms is increasing, the need for accurate recommendation systems is becoming increasingly important, especially in the animation industry. This research focuses on implementing a recommendation system that can help viewers easily navigate the abundance of content. By comparing collaborative filtering and content-based filtering methods, this research attempts to find the optimal approach for providing anime recommendations. From the results of A/B testing and further analysis, it was found that Collaborative Filtering was effective in providing recommendations based on similar interests between users. On the other hand, content-based filtering offers the advantage of personalizing recommendations based on content characteristics. Additionally, integrating these techniques into mobile applications will enrich the user experience, allowing them to receive recommendations more quickly and interactively. With these findings, this research contributes to the development of more intuitive and responsive recommendation systems, driving the growth of the anime streaming industry by increasing user satisfaction and retention.