Multi-Classifier Combination for Banks Credit Risk Assessment
Qifeng Zhou, Chengde Lin, Wei Yang · 2006
Credit risk assessment problem belongs essentially to a classification problem. In this paper, a multi-classifier combination algorithm has been developed for banks credit risk assessment. We adopt back-propagation (BP) algorithm as the meta-learning algorithm and compared the methods of bagging and boosting to construct the multi-classifier system (MCS). Experimental results on real client's data illustrate the effectiveness of the proposed method