Artificial intelligence tools such as expert systems and artificial neural systems have been heralded as new approaches to financial decision making. Expert systems, however, are unable to cope with incomplete or inaccurate data and are dependent on two value logic. Furthermore, extracting knowledge from the experts may be a costly and time consuming task. Artificial neural systems suffer from their inability to explain the steps by which they reach decisions and their inability to incorporate rules into their structure.
Fuzzy neural systems (FNS) address some of the short comings of artificial intelligence tools. An FNS uses a new concept called neural gates, which are similar to the processing elements in artificial neural systems but able to handle a broader range of information. Neural gates may be applied to forecasting stock market returns, assessing country risk and rating stocks based on fuzzy rules and probabilistic and Boolean data.