Universiti Teknologi Malaysia Institutional Repository

An economic forecasting based on association rules and neural network.

Defit, Sarjon and Md. Sap, Mohd. Noor (2001) An economic forecasting based on association rules and neural network. Jurnal Teknologi Maklumat, 13 (1). pp. 42-54. ISSN 0128-3790

[img]
Preview
PDF
918Kb

Abstract

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.

Item Type:Article
Uncontrolled Keywords:economic forecasting, neural networks, data mining, association rules
Subjects:H Social Sciences > HB Economic Theory
Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions:Computer Science and Information System (Formerly known)
ID Code:8761
Deposited By: Ms Zalinda Shuratman
Deposited On:12 May 2009 02:58
Last Modified:02 Jun 2010 01:57

Repository Staff Only: item control page