Jaccard with Singular Value Decomposition Hybrid Recommendation Algorithm

Abstract

In this paper, a comprehensive recommendation algorithm is studied, and a combination of Jaccard and singular value decomposition algorithm is proposed to improve the recommendation accuracy and recall rate. First, we use the Jaccard algorithm to find their recommended listings ranking matrix, which joined the incremental update. Then we use SVD algorithm to complete Jaccard matrix ranking algorithm, so as to obtain a complete list of recommendations, which are recommended according to actual requirements. Experimental results show that the proposed algorithm has higher performance compared with other related recommendation algorithms.