Robustness analysis of privacy-preserving model-based recommendation schemes

作者：

Highlights：

• We examine the robustness of model-based recommendation methods with privacy.

• SVD-based scheme with privacy is the most robust method against shilling attacks.

• Model-based prediction methods with privacy are more robust than memory-based ones.

• Segment attack is the most effective one against model-based schemes with privacy.

• Increasing filler size is more effective than increasing attack size.

摘要

•We examine the robustness of model-based recommendation methods with privacy.•SVD-based scheme with privacy is the most robust method against shilling attacks.•Model-based prediction methods with privacy are more robust than memory-based ones.•Segment attack is the most effective one against model-based schemes with privacy.•Increasing filler size is more effective than increasing attack size.

论文关键词：Robustness,Shilling,Privacy,Recommendation,Model,Collaborative filtering

论文评审过程：Available online 10 December 2013.

论文官网地址：https://doi.org/10.1016/j.eswa.2013.11.039