Integrating Multiple Recommendation Techniques for Enhanced Anime Discovery

Anime recommendation systems play a pivotal role in assisting users in discovering new content tailored to their preferences. This paper presents an innovative approach to anime recommendation utilizing content-based filtering techniques augmented by ensemble learning algorithms. The system leverages the TfIdfVectorizer to analyze textual data from anime names and ratings, extracting features to represent the content. Subsequently, ensemble learning algorithms including XGBoost, LightGBM, Gradient Boosting Decision Trees, AdaBoost, and CatBoost are employed to predict the accuracy of recommendations. The integration of content-based filtering and ensemble learning ensures a robust and accurate recommendation framework, enhancing user satisfaction and engagement. Experimental results demonstrate the effectiveness of the proposed system in providing personalized anime recommendations, thereby facilitating users in exploring diverse and appealing content within the anime domain.