Intelligent decision support aims at providing decision makers with tools that posses features usually associated with intelligent behaviour such as learning, reasoning, and uncertainty management. The paper describes a hybrid framework for this kind of intelligent DSS. In this framework neural networks and rough sets are integrated into a hybrid system and used cooperatively during the system lifecycle. The framework was applied in the building of a decision support system for issuing smog alerts in Melbourne. The data in the smog event forecasting database is noisy, imprecise, and often contradictory. Rough sets and neural networks were chosen for this application because they can discover patterns in ambiguous and imperfect data and provide tools for data and pattern analysis. The significance of the input features of the existing forecasting database was analysed using rough sets and subsequently predictive models based on these features were built and tested.
[1]
Z. Pawlak.
Rough Sets: Theoretical Aspects of Reasoning about Data
,
1991
.
[2]
William E. Spangler,et al.
The Role of Artificial Intelligence in Understanding the Strategic Decision-Making Process
,
1991,
IEEE Trans. Knowl. Data Eng..
[3]
C StylianouAnthony,et al.
Expert support systems
,
1993
.
[4]
Ilona Jagielska.
Rough Sets and Other Intelligent Techniques for Knowledge Discovery
,
1997,
ICONIP.
[5]
Jerzy W. Grzymala-Busse,et al.
Rough Sets
,
1995,
Commun. ACM.
[6]
Anthony C. Stylianou,et al.
Expert support systems: integrating AI technologies
,
1993,
CACM.