Intelligent Credit Risk Decision Support: Architecture and Implementations

Abstract

Recent breakthroughs in technology relevant to data analytics delivered many challenging opportunities for financial stakeholders. Continuously developed and applied novel artificial intelligence and machine learning techniques were proved to overcome the limitations of traditional statistical techniques and result in superior performance results. The leading financial rating companies, such as, Standard and Poor’s and Moody’s, are known to apply these techniques for practical rating of companies and financial instruments. Considering the growing importance of such techniques in industry and related research, this chapter attempts to present a review of such techniques applied to analyze insolvency, financial problems, develop rating systems and make financial decisions. Expert systems (ES) and decision support systems (DSS) are other important topics as their development involves many issues including consideration of which financial data should be used to implement such systems, the availability, integrity and quality. Several standards, such as XBRL (Extensible Business Language) or SDMX (Statistical Data and Metadata eXchange), provide means to solve these issues by ensuring relevant characteristics. As at the time of writing no modern DSS framework is known to propose such functionality, this chapter also presents an example framework for such financial standards-based DSS for credit risk evaluation, with respect to its architecture and implementation topics.