Trust-Centric and Attack-Resistant Recommender System
摘要
In the information age where social mediaSocial media and onlineOnline platforms are popular, recommendationRecommendation systems are being used more and more widely. As a result, the evaluation of recommendationRecommendation systems has become more rigorous, and a better-quality recommendationRecommendation system can largely improve the competitiveness of products. This chapter proposes three different ways to improve the performance of recommendationRecommendation systems based on both attackAttack and trustTrust: usage of the Simulated AnnealingSimulated Annealing AlgorithmAlgorithm in the score propagationScore propagation model, evaluation and usage of users’ activeness, and some administrative measuresAdministrative measures.