Please login to be able to save your searches and receive alerts for new content matching your search criteria.
Since the global financial crisis of 2008, it has been common sense that it is significant to identify and evaluate the risk to survive in the banking industry. The goal is to achieve a more accurate and efficient method of managing bank credit risk. Weighted k-nearest neighbor (k-NN) with information entropy is proposed, and real data is utilized to conduct comparison experiments with other different models and algorithms. As the result shows, weighted k-NN with information entropy is the best tool in contrast with other models. The method of weight k-NN with entropy could be applied in reality.