Data Mining Techniques for Credit Risk Assessment Task
This paper is review of current usage of data mining, machine learning and other algorithms for credit risk assessment. We are witnessing importance of credit risk assessment, especially after the global economic crisis on 2008.S o, it is very important to have a proper way to deal with the credit risk and provide powerful and accurate model for credit risk assessment. Many credit scoring techniques such as statistical techniques (logistic regression, discriminant analysis) or advanced techniques such as neural networks, decision trees, genetic algorithm, or support vector machines are used for credit risk assessment. Some of them are described in this article with theirs advantages/disadvantages. Even with many models and methods, it is still hard to say which model is the best or which classifier or which data mining technique is the best. Each model depends on particular data set or attributes set, so it is very important to develop flexible model which is adaptable to every dataset or attribute set.