An Improved Collaborative Filtering Recommendation Algorithm

With the inflationary development of data sizes, the traditional recommendation system the recommendation efficiency of the recommendation system encounters a great challenge. This paper proposes an improved collaborative filtering algorithm. The algorithm firstly implements the clustering operation data, takes average rating and evaluated times of item as dimension of two dimensional grids, and then implements clustering based on item with CLIQUE grid clustering algorithm in accordance with item similarity. The corresponding experiment shows the method can significantly improve the recommendation efficiency of the recommendation system.