An Effective Collaborative User Model Using Hybrid Clustering Recommendation Methods

Collaborative Filtering (CF) has been known as the most successful recommendation technique in which recommendations are made based on past rating records from like-minded users. Significant growth of users and items have negatively affected efficiency CF pose key issues related to computational aspects quality such high dimensionality data sparsity. In this study, a hybrid method was proposed capable solve mentioned problems using neighborhood selection process for each user through two clustering algorithms were item-based k-means user-based Fuzzy Clustering. Item-based applied because its advantages time hence it is able address issues. To create groups find correlation between users, we employed Clustering not yet used clustering. This can calculate degree membership among into set clustered items. Furthermore, new similarity metric designed compute value with affecting output an alternative basic metrics proven provide high-quality noticeable improvement accuracy The evaluated benchmark datasets, MovieLens LastFM order make comparison existing methods.