Application of the K-Nearest Neighbors Algorithm in Anime Recommendation System Based on Genre Preferences and Web-Based Ratings

The rapid growth of the anime industry has resulted in an increasing number of anime titles being released every year. The large variety of available anime often makes it difficult for users to find anime that match their interests and preferences. The process of searching for anime manually requires considerable time, as users need to review genres and ratings individually. Therefore, a recommendation system is needed to assist users in finding suitable anime more quickly and efficiently. This study aims to develop a web-based anime recommendation system using the KNearest Neighbors (KNN) algorithm based on users' genre and rating preferences. The dataset used in this research was obtained from the Anime Recommendations Database available on Kaggle, consisting of 12,294 anime records. The research process includes data collection, data cleaning, attribute selection, data transformation, KNN algorithm implementation, and web-based system development. The KNN algorithm is applied to calculate the similarity between user preferences and anime data using the Euclidean Distance method. The results of this study indicate that the developed system is capable of providing anime recommendations that match users' preferences based on selected genres and ratings. The system also provides features for anime data management, recommendation searches, anime detail views, and a watchlist feature for saving anime that users intend to watch. The implementation of this system helps users find suitable anime more easily, quickly, and efficiently compared to manual searching methods. .