Research map / Architectures and efficiency / RNNs and normalization
Exploiting Data Sparsity in Secure Cross-Platform Social Recommendation
Jamie Cui, Chaochao Chen, Lingjuan Lyu, Carl Yang, Wang Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use social data from the social platform (Soreg/S3Rec) instead of rating-only MF | matrix factorization · Epinions | RMSE: 1.193 -> 1.062 | Improved |
| use social data instead of rating-only MF | matrix factorization · LibraryThing | RMSE: 0.927 -> 0.908 | Improved |
| use S3Rec secure sparse protocols instead of SeSoRec | secure social recommendation · Epinions | RMSE: 1.062 -> 1.062 | No clear effect |
| use S3Rec sparse secure matrix multiplication instead of SeSoRec | secure social recommendation · Epinions | total time per epoch (s): 7,799 -> 419.9 | Improved |
| use S3Rec instead of SeSoRec | secure social recommendation · LibraryThing | total time per epoch (s): 16,084 -> 262.1 | Improved |
| use S3Rec instead of SeSoRec | secure social recommendation · LibraryThing | total communication (GB): 1,365 -> 2.201 | Improved |
| use S3Rec instead of SeSoRec | secure social recommendation · Epinions | total communication (GB): 798.6 -> 3.552 | Improved |
| reduce social data density by sampling 0.8, 0.6, 0.4 of relations (ablation) | S3Rec · Epinions | total time (s): 401.8 (0.8) -> 366.3 (0.4) | Improved |
| reduce social data density by sampling (ablation) | S3Rec · Epinions | RMSE: 1.0932 -> 1.1751 | Hurt |
| increase latent dimension k from 10 to 20 (ablation) | S3Rec · Epinions | total time (s): 419.9 -> 527.4 | Hurt |
| increase latent dimension k from 10 to 20 (ablation) | S3Rec · Epinions | total communication (GB): 3.552 -> 3.552 | No clear effect |
Closest papers by meaning
- SecEmb: Sparsity-Aware Secure Federated Learning of On-Device Recommender System with Large Embedding (ICML 2025 · similarity 0.86)
- Rethinking Byzantine Robustness in Federated Recommendation from Sparse Aggregation Perspective (AAAI 2025 · similarity 0.83)
- Reducing Item Discrepancy via Differentially Private Robust Embedding Alignment for Privacy-Preserving Cross Domain Recommendation (ICML 2024 · similarity 0.83)
- RRL: Recommendation Reverse Learning (AAAI 2024 · similarity 0.81)
- Federated Graph Learning for Cross-Domain Recommendation (NeurIPS 2024 · similarity 0.81)
- CURE4Rec: A Benchmark for Recommendation Unlearning with Deeper Influence (NeurIPS 2024 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
