Influence and Correlation in Social Networks

Abstract:

In many social systems, social ties between users play an important role in dictating users' behavior. One of the ways this can happen is through social influence, the phenomenon that the actions of a user can induce his/her friends to behave in a similar way. Identifying and understanding social influence is of tremendous interest from both an analysis (e.g., predicting the future of the system) and a design (e.g., designing viral marketing strategies) point of view. This is a difficult task in general, since there are many other factors that can induce statistical correlation between the actions of friends in a social network. In this paper, we propose two simple tests that can identify influence as a source of social correlation in cases where data on the time step of actions are available. We give a theoretical justification of one of the tests by proving that with high probability it succeeds in ruling out influence in a rather general model of social correlation. We also simulate our tests on a number of examples designed by randomly generating actions of nodes on a real social network (from Flickr) according to one of several models. Finally, we apply them to real tagging data on Flickr, exhibiting that while there is significant social correlation in tagging behavior on this system, this correlation cannot be attributed to social influence.In many social systems, social ties between users play an important role in dictating users' behavior. One of the ways this can happen is through social influence, the phenomenon that the actions of a user can induce his/her friends to behave in a similar way. Identifying and understanding social influence is of tremendous interest from both an analysis (e.g., predicting the future of the system) and a design (e.g., designing viral marketing strategies) point of view. This is a difficult task in general, since there are many other factors that can induce statistical correlation between the actions of friends in a social network. In this paper, we propose two simple tests that can identify influence as a source of social correlation in cases where data on the time step of actions are available. We give a theoretical justification of one of the tests by proving that with high probability it succeeds in ruling out influence in a rather general model of social correlation. We also simulate our tests on a number of examples designed by randomly generating actions of nodes on a real social network (from Flickr) according to one of several models. Finally, we apply them to real tagging data on Flickr, exhibiting that while there is significant social correlation in tagging behavior on this system, this correlation cannot be attributed to social influence.