An economic forecasting based on association rules and neural network.

• Main Authors: Defit, Sarjon, Md. Sap, Mohd. Noor
• Format: Article
• Language: English
• Published: Penerbit UTM Press 2001
• Subjects: HB Economic Theory; QA75 Electronic computers. Computer science
• Online Access: http://eprints.utm.my/id/eprint/8761/1/MohdNoorMd2001_AnEconomicForecastingBasedOn.PDF; http://eprints.utm.my/id/eprint/8761/
• Institution: Universiti Teknologi Malaysia

The world economic environment is currently undergoing drastic changes, such as economic forecasting. It is a complex and challenging task because of the following reasons: i) there is no economic model which carries conviction, ii) economic time series are intrinsically very unreliable and generally have poor signal to accurate results, iii) non stationary and iv) non linearity. In this paper we propose an economic forecasting based on association rules and neural network. Our study concludes that our model can improve the prediction ability by using frequent itemsets and predicted attribute as input fields. In this paper, we explain the important of data mining and prediction, association rule mining, the proposed economic forecasting model, testing and experimental results, model performance and conclusion.