The following paper has been accepted at the 13th ACM Conference on Recommender Systems (RecSys’19) [1]:

“Efficient Privacy-Preserving Recommendations based on Social Graphs”

(Aidmar Wainakh, Tim Grube, Jörg Daubert, Max Mühlhäuser)

The paper tackles the efficiency problem of privacy-preserving association rules mining (PPARM) for distributed data in online social networks.

The authors propose using the social graph as a base for sampling the data prior to the mining process.

By that, they achieve improvements in efficiency and privacy of PPARM approaches.

[1] https://recsys.acm.org/recsys19/