Personal credit scoring is a fundamental problem in Finance. It has a lot of emerging applications including Credit Card,Mortgage Loan and Automobile Credit, etc. Due to its importance , a lot of personal credit scoring methods have been proposed. Among many other aspects of a client,trade reference is a crucial aspect for an effective credit scoring. In a trade reference,the referee provides a score to describe the credit of the client with respect to the business between the client and the referee. Most existing methods assume that the trade reference is trustable. However,the trade referee may conspire with the client and the trade reference score becomes untrustable in practice. Therefore, in this paper, we propose a trade referee rank to capture both the reputation of the trade referee and the trade reference score provided by the referee. The accuracy of using our trade reference rank is about 40% higher than that of the existing methods. We propose a random walk based method on a client reference graph to compute the trade reference rank of the clients. Our extensive experiments techniques.
展开▼