Development of a Web-Based Anime Streaming and Recommending Application: Enhancing User Experience with Naive Bayes Classification and Floating Text

This project demonstrates the application of Naive Bayes Classification to develop an anime recommender system, using content-based filtering approach. The project utilizes genre data from a comprehensive anime database obtained from online source(s)  to train the model in categorizing anime and displaying recommendations to users. The findings indicate that calculating recommendations based on the user's most recent three watched anime provides the most accurate results. This approach enhances flexibility and variety in recommendations, reducing repetitive suggestions and expanding the range of genres. Additionally, the project incorporates a floating text feature, allowing real-time user interactions and comments, fostering a sense of anime community engagement among anime enthusiasts. This system is expected to beneficial to anime fans by enhancing viewing experiences and fostering stronger community engagement. The study concludes that integrating interactive features like floating text is vital for enhancing the more engagement that leads to the better user experience (UX).