Visible to the public Research of Personalized Recommendation Algorithm Based on Trust and User's Interest

TitleResearch of Personalized Recommendation Algorithm Based on Trust and User's Interest
Publication TypeConference Paper
Year of Publication2018
AuthorsSun, P., Yin, S., Man, W., Tao, T.
Conference Name2018 International Conference on Robots Intelligent System (ICRIS)
KeywordsAccuracy, Clustering algorithms, Collaboration, filtering algorithms, History, Human Behavior, human factors, joint interest-content recommendation, Motion pictures, online video site, pattern clustering, personalized recommendation algorithm, Prediction algorithms, pubcrawl, recommender systems, resilience, Resiliency, Robot Trust, robust trust, social networking (online), sparse subspace clust algorithm, Trust, trust similarity, Trusted Computing, trustworthyness model, user interest, Videos
Abstract

Most traditional recommendation algorithms only consider the binary relationship between users and projects, these can basically be converted into score prediction problems. But most of these algorithms ignore the users's interests, potential work factors or the other social factors of the recommending products. In this paper, based on the existing trustworthyness model and similarity measure, we puts forward the concept of trust similarity and design a joint interest-content recommendation framework to suggest users which videos to watch in the online video site. In this framework, we first analyze the user's viewing history records, tags and establish the user's interest characteristic vector. Then, based on the updated vector, users should be clustered by sparse subspace clust algorithm, which can improve the efficiency of the algorithm. We certainly improve the calculation of similarity to help users find better neighbors. Finally we conduct experiments using real traces from Tencent Weibo and Youku to verify our method and evaluate its performance. The results demonstrate the effectiveness of our approach and show that our approach can substantially improve the recommendation accuracy.

DOI10.1109/ICRIS.2018.00048
Citation Keysun_research_2018