Abstract
Nowadays with the readily accessibility of online social networks (OSNs), people are facilitated to share interesting information with friends through OSNs. Undoubtedly these sharing activities make our life more fantastic. However, meanwhile one challenge we have to face is information overload that we do not have enough time to review all of the content broadcasted through OSNs. So we need to have a mechanism to help users recognize interesting items from a large pool of content. In this project, we aim at filtering unwanted content based on the strength of trust relationships between users. We have proposed two kinds of trust models — basic trust model and source-level trust model. The trust values are estimated based on historical user interactions and profile similarity. We estimate dynamic trusts and analyze the evolution of trust relationships over dates. We also incorporate the auxiliary causes of interactions to moderate the noisy effect of user's intrinsic tendency to perform a certain type of interaction. In addition, since the trustworthiness of diverse information sources are rather distinct, we further estimate trust values at source-level. Our recommender systems utilize several types of Collaborative Filtering (CF) approaches, including conventional CF (namely user-based, item-based, singular value decomposition (SVD)-based), and also trust-combined user-based CF. We evaluate our trust models and recommender systems on Friendfeed datasets. By comparing the evaluation results, we found that the recommendations based on estimated trust relationships were better than conventional CF recommendations.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.