Abstract

This study proposes a credit scoring model for SMEs using transactional data–based variables. Offline and online test results both confirmed the predictive capability of transactional data–based variables. Furthermore, to align with business objectives, a two-stage multi-objective feature-selection method that optimizing the model profit and model risk is proposed for model deployment. The results showed that the proposed method greatly reduced the cardinality of the feature subset and therefore improved model comprehensibility. What’s more, the proposed method provides a trade-off between profit and risk for banks, as they can choose an appropriate feature subset based on their risk preference.

About the Speaker

Prof. Gang Kou is currently the Executive Dean of School of Business Administration, and the Dean of Research Institute of Big Data, Southwestern University of Finance and Economics (SWUFE). He is the member of National MBA Education Steering Committee of China, a Distinguished Professor of Chang Jiang Scholars Program (MOE of China), the managing editor of International Journal of Information Technology & Decision Making (SCI), and the managing editor-in-chief of Financial Innovation (SSCI). He is also editors for the following journals: Decision Support Systems, European Journal of Operational Research, Technological and Economic Development of Economy. Previously, he was a professor of School of Management and Economics, University of Electronic Science and Technology of China, and a research scientist in Thomson Co., R&D. He received his Ph.D. in Information Technology from the College of Information Science & Technology, Univ. of Nebraska at Omaha; Master degree in Department of Computer Science, Univ. of Nebraska at Omaha; and B.S. degree in Department of Physics, Tsinghua University, China. He has published more than 100 papers in various peer-reviewed journals and his papers have been cited for more than 9000 times.