Covert nodes mining in social networks based on games theory

The problem of discovering covert nodes in social network has been widely studied because of its tremendous number of applications in determining critical points in social network, such as detecting terrorist, recommending item for possible customer, finding the source of spreading gossip, etc. In this paper, we utilize game theory to solve this problem. Firstly, we propose the model which analyzes game in the influence transmission. Then we obtain each nodes contribution by calculating the nodes earning in game. The feasibility and effectiveness of our method were verified on a simulation dataset and a real dataset.