A Lasso-based Collaborative Filtering Recommendation Model

This paper proposes a new approach to solve the problem of lack information in rating data due users or items, there is too little user for items collaborative filtering recommendation models (CFR models). In this approach, we consider similarity between based on lasso regression build CFR models. commonly used models, results are built only feedback matrix users. The our model predicted two calculated values: (1) value matrix; (2) prediction Lasso regression. experimental proposed popular datasets have been processed and integrated into recommenderlab package showed that suggested higher accuracy than result confirms helps deal with