FCT: a fully-distributed context-aware trust model for location based service recommendation

摘要: With the popularity of location based service (LBS), a vast number trust models for LBS recommendation (LBSR) have been proposed. These are centralized in essence, and trusted third party may collude with malicious providers or cause single-point failure problem. This work improves classic certified reputation (CR) model proposes novel fully-distributed context-aware (FCT) LBSR. Recommendation operations conducted by directly is no longer required our FCT model. Besides, also supports movements due to its self-certified characteristic. Moreover, easing collusion attack value imbalance attack, we comprehensively consider four kinds factor weights, namely number, time decay, preference context weights. Finally, scenario deployed, comprehensive experiments analysis conducted. The results indicate that significantly outperforms CR terms robustness against as well performance improving successful trading rates honest reducing risks providers.