APPLICATION OF CLUSTERING METHODS IN RECOMMENDER SYSTEMS FOR USER BEHAVIOR ANALYSIS

Recommender systems are crucial in personalizing digital experiences by predicting user preferences. This paper examines the application of the k-means clustering algorithm in recommender systems to segment users based on behavioural patterns, enhancing recommendation accuracy and efficiency. The study also compares k-means with other clustering techniques, analyzing their advantages and limitations in handling sparsity and the cold start problem. The results highlight the effectiveness of k-means for improving user segmentation and optimizing recommendation strategies.