Make best use of social networks via more valuable friend recommendations

The human factors behind how a user gets in touch with the others are complex especially in twitter-like social networks, which unlike Facebook-like social networks, are gradually showing great power in information propagation. In an effective friend recommendation, the identification of factors that influence links creation between users is essential. This paper makes a full study of the human factors in social networks and takes into account both the users' need of similar friends and diversified friends. This paper, focusing on Twitter-like social networks, enumerates several of those intuitive and connotative criteria in establishing friendship on-line and then designs a recommendation system that fit Twitter-like social networks to help improve the user experience and help user benefit from the architecture and resources from social networks. The recommendation mechanism is developed based on the incorporation of heterophily value and homophily value in establishing friendship into hybrid content and collaborative filtering recommendation algorithm.