Over the past decade, Decision Support Systems have become very popular and they have been used in a wide variety of applications. More recently, efforts have been concentrating on using them for strategic decision-making. The authors are currently developing a Strategic Decision Support System (SDSS) for consulting engineering firms. This is being achieved by combining the two popular technologies of neural networks and expert systems to create hybrid system. This paper focuses on the problem that often presents itself when training a neural network, namely the lack of adequate training sets, and proposes an innovative approach for overcoming this problem. A method for artificially generating the training sets is outlined. and its use in training the neural network is described. Finally, the usefulness of this simulated set of training data for testing the validity and robustness of the SDSS is assessed, followed by a discussion on how this approach can be extended to other applications.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £55 +P&P)