A paper jointly written by NOVA IMS PhD student Afshin Ashofteh and Jorge Bravo, Professor at NOVA IMS presented at the 8th International Conference on Risk Analysis and Design of Experiments (2019) in Vienna has won the first prize for the best paper presented at the Conference.
This paper aimed at the case of credit scoring for risk management. It presented a novel method for credit scoring to be used for default prediction. Financial institutions and lenders use credit scores to assess the creditworthiness of new clients and to decide whether to grant credit or approve a loan. Recently, mobile phones provide data sources in the form of call-detail records (CDR) and many other log files. These new sources of Big Data attain much more importance to provide smarter credit-scoring models not only for customers of banks but also for largely unbanked population who has no regular credit history. Therefore, this new credit scoring methodology represents a machine-learning algorithm for Big Data sources, which is less expensive in computation with higher accuracy in compare with current algorithms.
The ICRA conferences are organized every two years by the ISI Committee on Risk Analysis (ISI-CRA), one of the nine Special Interest Groups of the ISI. This year’s conference focuses on Risk Analysis and Experimental Design and it was devoted to the Distinguished Professor Samad Hedayat from UIC-USA.