Intelligent Link Prediction Management Based on Community Discovery and User Behavior Preference in Online Social Networks

Link prediction in online social networks intends to predict users who are yet to establish their network of friends, with the motivation of offering friend recommendation based on the current network structure and the attributes of nodes. However, many existing link prediction methods do not consider important information such as community characteristics, text information, and growth mechanism. In this paper, we propose an intelligent data management mechanism based on relationship strength according to the characteristics of social networks for achieving a reliable prediction in online social networks. Secondly, by considering the network structure attributes and interest preference of users as important factors affecting the link prediction process in online social networks, we propose further improvements in the prediction process by designing a friend recommendation model with a novel incorporation of the relationship information and interest preference characteristics of users into the community detection algorithm. Finally, extensive experiments conducted on a Twitter dataset demonstrate the effectiveness of our proposed models in both dynamic community detection and link prediction.

A Potential Friend Recommendation Algorithm for Obtaining Spatial Information

With the rapid development of social network, friend recommendation algorithm has become an important component of social application. Location-based social network (LBSN) enables users to record and share their locations anytime and anywhere, which is a high quality information source. In order to meet people's demand of expanding social circle and obtaining diversified spatial information when making friends, this paper proposes a potential friend recommendation algorithm based on the similarity of user's check-in behavior and spatial information acquisition level in the real world. Firstly, we employ kernel density estimation and time entropy to solve the problems of data sparsity and low concentration, then employ cosine distance to measure the check-in behavior similarity. Secondly, we analyze users’ spatial distribution of check-in location and cognitive differences on spatial information. Finally, the method mentioned above is tested with dataset called Foursquare. The results of the experiment show that the proposed method has competitive performance.

Social Recommendation Model with User Trust and Item Ratings Using Collaborative Filtering Technique in Hotel Application

A trust-based recommendation model is regularized with user trust and item ratings called TrustSVD. Trust networks are large-world networks where many users are socially linked, suggesting the assumption of trust in recommendation systems. An item rating downloaded from the OSN Server can be viewed by the user. If the information is accessible on the server, all the adjacent devices are enabled and a peer to peer mode of communication is initiated. User reviews from a graphical forum are shown. It focuses on the rating prediction role in the current framework and has shown that integrating user social confidence data will boost the output of recommendations. The strategy builds on the SVD++ state-of-the-art model. The data sparsity and cold start issues are resolved in the friend of friend recommendation model used. The mining method generates the user's overall rating in graphical representations and illustrates the overall rating. This model increases the utility of data by exchanging neighborhoods to protect security and privacy issues. One of the most common techniques for implementing a recommendation scheme is Collaborative filtering (CF).