The maximum imputation framework for neighborhood-based collaborative filtering

Yongli Ren, Gang Li, Jun Zhang, Wanlei Zhou

Abstract

The data sparsity is a well-known issue in the context of collaborative filtering, and it puts particular difficulties in making accurate recommendations. In this paper, we focus on the data sparsity problem in the context of neighborhood-based collaborative filtering, and propose a maximum imputation framework to tackle this. The basic idea is to identify an imputation area that can maximize the imputation benefit for recommendation purposes, while minimizing the imputation error brought in. To achieve the maximum imputation benefit, the imputation area is determined from both the user and the item perspectives; to minimize the imputation error, there is at least one real rating preserved for each item in the identified imputation area. A theoretical analysis is provided to prove that the proposed imputation method outperforms the conventional neighborhood-based CF methods through more accurate neighbor identification. We evaluate the proposed framework on two benchmark datasets by comparing it with seven relevant methods. The experimental results indicate that the proposed method significantly outperforms other relevant methods.
BibTeX
@article{ren2014maximum,
  author = {Yongli Ren and Gang Li and Jun Zhang and Wanlei Zhou},
  title = {{The maximum imputation framework for neighborhood-based collaborative filtering}},
  journal = {Soc. Netw. Anal. Min.},
  year = {2014},
  doi = {10.1007/s13278-014-0207-3},
}

Loading related content...

Figure