Research Repository

An Abstracted 'Not so Expert' System

McGrath, Michael (2009) An Abstracted 'Not so Expert' System. In: 4th IASTED International Conference on Computational Intelligence 2009. Kovalerchuk, B, ed. Curran Associates, Red Hook, NY, pp. 15-22.

Full text for this resource is not available from the Research Repository.

Abstract

After falling out of favour during the 1990s, advisory expert systems and their underlying artificial intelligence (AI) technologies are now being used increasingly in organizational decision making activities – particularly, data mining. In this paper, an expert system designed for greyhound race tipping specifically (and investment advice more generally) is described. Our system is distinguished from others in two ways: i) the use of an abstracted schema allows ready extension to other investment domains; and ii) emphasis is placed on reducing complexity where possible – hence, the ‘not so expert’ system. Early indications are that the forecasts produced by our system compare favourably with those of both human experts and previous, relevant decision support systems (DSS). In: Proceedings Of The 4th IASTED International Conference on Computational Intelligence held 17-19 August 2009, Honolulu, Hawaii, USA

Item Type: Book Section
ISBN: 9781615677528 (print) 9780889868069 (online)
Uncontrolled Keywords: ResPubID16938, expert systems, data abstraction, betting
Subjects: Faculty/School/Research Centre/Department > School of Management and Information Systems
FOR Classification > 0806 Information Systems
SEO Classification > 9003 Tourism
Related URLs:
Depositing User: VUIR
Date Deposited: 02 Dec 2013 03:45
Last Modified: 02 Dec 2013 03:45
URI: http://vuir.vu.edu.au/id/eprint/5143
ePrint Statistics: View download statistics for this item
Citations in Scopus: 1 - View on Scopus

Repository staff only

View Item View Item

Search Google Scholar