An imputation-based matrix factorization method for improving accuracy of collaborative filtering systems

Manizheh Ranjbar, Parham Moradi, Mostafa Azami, Mahdi Jalili

Abstract: Highlights•A novel matrix factorization method called IMULT is proposed.•IMULT utilizes imputed rating to mitigate the sparsity problem.•The convergence proof of IMULT update rules is provided.•The results show that IMULT performed better than ALS, SGD, RSGD, SVD++ and MULT.