As online social networks have become part of daily life, social aware recommendation has progressively emerged, aiming at more sophisticated schemata to support automated recommendations.

Current solutions are based on various models that reflect significantly different approaches. Most of them do not take into consideration “the interpersonal context-sensitive trust that exists between individuals in [a] social network.” Trust intrinsically affects aspects of social interaction, including the way recommendations are made and interpreted. In this paper, the authors propose a privacy-preserving trusted social feedback (TSF) schema where users can obtain feedback from their social network. The concept can be extended to deal with crowdsourcing, where feedback is retrieved from experts of a given domain. The schema is designed to support “categorical answers as well as single-valued numerical answers.” In order to provide a consistent solution to trust propagation, the authors have associated trust with the strength of a social relation, postulating that it is “one’s asymmetric personal perception of another in a particular context that changes over time.”

The paper is interesting, clear, and well written. Furthermore, feedback from a "user study to evaluate the perception of privacy and foreground trust in the prototype" shows promising results. I recommend reading it.