Personalized Privacy-Preserving Social Recommendation
Authors: Xuying Meng, Suhang Wang, Kai Shu, Jundong Li, Bo Chen, Huan Liu, Yujun Zhang
AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this section, we conduct experimental evaluation to validate the effectiveness of Priv SR. We aim to answer two questions: (1) can Priv SR improve recommendation effectiveness by incorporating sensitive ratings and social relationships? and (2) can it protect sensitive ratings under reconstruction attack while retaining recommendation effectiveness? In the following, we first introduce our datasets and experimental settings, and then conduct experimental evaluation followed by analyzing impacts of parameters. |
| Researcher Affiliation | Academia | Xuying Meng,1,2 Suhang Wang,3 Kai Shu,3 Jundong Li,3 Bo Chen,4 Huan Liu,3 Yujun Zhang1 1Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China 2University of Chinese Academy of Sciences, Beijing 100049, China 3Computer Science and Engineering, Arizona State University, Tempe, 85281, USA 4Department of Computer Science, Michigan Technological University, Houghton, 49931, USA |
| Pseudocode | Yes | Algorithm 1 Priv SR Algorithm |
| Open Source Code | Yes | Meng, X.; Wang, S.; Shu, K.; Jundong, L.; Chen, B.; Liu, H.; and Zhang, Y. 2018b. Personalized privacy-preserving social recommendation. https://github.com/mxyenguing/ Priv SR/blob/master/appendix.pdf. |
| Open Datasets | Yes | Two publicly available datasets Ciao5 and Epinions6 are used for evaluation. For both datasets, users can rate products from 1 to 5 and establish social relations with others. Detailed statistics of these two datasets are shown in Table 1. 5http://www.ciao.co.uk/ 6http://www.epinions.com/ |
| Dataset Splits | Yes | We use five-fold cross validation for the following experiments. |
| Hardware Specification | No | No specific hardware details are mentioned in the paper. |
| Software Dependencies | No | The paper does not mention any software with specific version numbers. |
| Experiment Setup | Yes | We then set γ = 10 4, λ = 10 3, α = 10 2 and the dimension K = 10. |