A differential privacy framework for collaborative filtering
Summary: Focusing on the privacy issues in recommender systems, we propose a framework containing two perturbation methods for differentially private collaborative filtering to prevent the threat of inference attacks against users. To conceal individual ratings and provide valuable predictions, we consider some representative algorithms to calculate the predicted scores and provide specific solutions for adding Laplace noise. The DPI (Differentially Private Input) method perturbs the original ratings, which can be followed by any recommendation algorithms. By contrast, the DPM (Differentially Private Manner) method is based on the original ratings, which perturbs the measurements during implementation of the algorithms and releases the predicted scores. The experimental results showed that both methods can provide valuable prediction results while guaranteeing DP, which suggests it is a feasible solution and can be competent to make private recommendations.
- On Differentially Private Online Collaborative Recommendation Systems
- New SVD-based collaborative filtering algorithms with differential privacy
- Privacy-preserving and yet robust collaborative filtering recommender as a service
- Locally differentially private item-based collaborative filtering
- New collaborative filtering algorithms based on SVD++ and differential privacy
- New collaborative filtering algorithms based on SVD++ and differential privacy
- Jo-DPMF: differentially private matrix factorization learning through joint optimization
- Locally differentially private item-based collaborative filtering
- A smart privacy-preserving learning method by fake gradients to protect users items in recommender systems
- On Differentially Private Online Collaborative Recommendation Systems
- New SVD-based collaborative filtering algorithms with differential privacy
- Privacy-preserving and yet robust collaborative filtering recommender as a service
- USST: a two-phase privacy-preserving framework for personalized recommendation with semi-distributed training
- A differentially private nonnegative matrix factorization for recommender system
This page was built for publication: A differential privacy framework for collaborative filtering
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2298067)
