On the model of credit score calculation using social networks data

The main factor of making a decision or solving a problem is having enough information; the more parameters are given, the easier the solution. The problem considered here is to determine whether to give a credit to an applicant or not. A lot of information about the applicant is needed in order to make this decision. This information can be obtained from the applicant himself or from the agencies specialized in such business. Now, the largest and renewable source of information about people is the social networks. They are an inexhaustible resource, and that is so valuable for advertising companies, banks and lenders, who usually suffer from a lack of data about applicants for a loan and cannot determine the credit scores accurately. In this paper, the authors present a few models that can extract social networks data and use them to predict the statues of an applicant based on the data on similar applicants. The data of social network VKontakte (http://www.vk.com/) are used.