A member recognition approach for specific organizations based on relationships among users in social networking Twitter.

It is very meaningful and also has good business value to map relationships among users from virtual network to reality. But the current research mainly concentrates on community discovery in social networks. To address the issue, we propose an approach that can mine colleagueship among users in social networking and then find out the members for specific organizations. In this paper, first we define 6 parameters for quantitatively describing the relationship between a user and a group of users on Twitter. And then, we define 7 hypotheses for describing the interactive features between colleagues on Twitter so as to apply the 6 parameters to colleagueship mining and member recognition. Then through empirical research we systematically evaluate the influence of each of the 6 parameters on identifying colleagueship on Twitter. Finally, we present an optimal evaluation model for our approach to determine whether a user is to be a member of a specific organization. Given an organization with its public account and a list of sample users on Twitter, our approach can dig out the users who affiliate with the organization. We also conduct an experiment to evaluate our approach. The experimental results demonstrate that our approach is superior to the main existing schemes in terms of a high recognition rate.