Abstract:
In a typical decision support environment, the decision-making process deploys asmall portion of the needed information that exists on computers and the structures of thedecision models are highly generalized for each of the problems. However, knowledge in the minds of the decision makers, called tacit knowledge, is needed to give convenient decisions and to construct specific models for problems. This situation coerces the development of anew decision support concept that integrates knowledge management and decision support systems by using knowledge discovery techniques. The purpose of the study is to develop a framework that applies knowledge discovery techniques for various types of knowledge conversion and generates specific decision models by utilizing previously defined models asmuch as possible. In order to prove the applicability of the proposed framework, anexperimental study is designed. The chosen problem domain forecasts the Turkish financialand macroeconomic time series. In addition to its main purpose, the study suggests to increase the effectiveness of decision support systems, to enhance the knowledge in the decision making process and the reby to improve the decision-making process.
