<![CDATA[This study aims to analyze anime audience preferences based on genres using the K-Means clustering algorithm. The dataset consists of 100 popular anime titles with features such as ratings, votes, and genres. The research steps include data preprocessing, clustering with the Elbow method to determine the optimal number of clusters, and applying the K-Means algorithm. The clustering results revealed four clusters with unique characteristics, highlighting differences in popularity and genre preferences. Evaluation using the Confusion Matrix shows a model accuracy of 95%, while the Silhouette score of 0.285 indicates adequate cluster separation. These findings are expected to provide insights for streaming platforms to deliver more personalized and relevant anime recommendations to viewers.]]>
Copyrights © 2025
