A Product Recommendation Algorithm Based on the Expectation of a Not Scored Item by the Prospective User

Author(s):  
Zheng Zhai ◽  
Yang-Jie Chu ◽  
Jia-Min Wei
2021 ◽  
Author(s):  
Jian-hua LI ◽  
Chen-xi ZHANG ◽  
Chun-li LEI ◽  
Hong ZHANG ◽  
Lin-long CHEN

2012 ◽  
Vol 461 ◽  
pp. 289-292
Author(s):  
Kai Zhou

Recommender systems are becoming increasingly popular, and collaborative filtering method is one of the most important technologies in recommender systems. The ability of recommender systems to make correct predictions is fundamentally determined by the quality and fittingness of the collaborative filtering that implements them. It is currently mainly used for business purposes such as product recommendation. Collaborative filtering has two types. One is user based collaborative filtering using the similarity between users to predict and the other is item based collaborative filtering using the similarity between items. Although both of them are successfully applied in wide regions, they suffer from a fundamental problem of data sparsity. This paper gives a personalized collaborative filtering recommendation algorithm combining the item rating similarity and the item classification similarity. This method can alleviate the data sparsity problem in the recommender systems


Author(s):  
K Sobha Rani

Collaborative filtering suffers from the problems of data sparsity and cold start, which dramatically degrade recommendation performance. To help resolve these issues, we propose TrustSVD, a trust-based matrix factorization technique. By analyzing the social trust data from four real-world data sets, we conclude that not only the explicit but also the implicit influence of both ratings and trust should be taken into consideration in a recommendation model. Hence, we build on top of a state-of-the-art recommendation algorithm SVD++ which inherently involves the explicit and implicit influence of rated items, by further incorporating both the explicit and implicit influence of trusted users on the prediction of items for an active user. To our knowledge, the work reported is the first to extend SVD++ with social trust information. Experimental results on the four data sets demonstrate that our approach TrustSVD achieves better accuracy than other ten counterparts, and can better handle the concerned issues.


Sign in / Sign up

Export Citation Format

Share Document