LINK PREDICTION IN SOCIAL NETWORK BY SNA AND SUPERVISED LEARNING

Social network Analysis research has attracted much attention in recent years. Link prediction is a key research direction within this area. Social network analysis (SNA) and supervised learning with classification algorithms are used in this research for the link prediction in social networking. SNA is used to analyze the key factors influencing information diffusion model about density, centrality and the cohesive subgroup, reveals useful insights which are the relationships and activities between human and identify the influencing nodes. The supervised learning with the classification algorithms of Independent Cascade Model [6] is used to focus on discovering surprising links in the existing ones of influencing nodes. The partitions store discrete characteristics of nodes. The supervised learning for prediction which nodes in a social network will be linked next from the attacked influencing nodes with the classification algorithms to monitoring the risk. Such understanding can lead to efficient implementation of tools to link prediction in social network. They are applied as a guide to further investigation of social network behaviors and improve the security model for social networking to preventing and controlling information diffusion or computer viruses in network application.