Symbolic Rule Extraction From Neural Networks: An Application to Identifying Organizations Adopting IT

Information & Management, 34, 2 (1998), pp. 91-101.

HKUST Business School Research Paper No. 2021-006

14 Pages Posted: 20 Mar 2021 Last revised: 4 Jun 2021

See all articles by Rudy Setiono

Rudy Setiono

National University of Singapore

James Y.L. Thong

HKUST Business School

Chee-Sing Yap

affiliation not provided to SSRN

Date Written: 1998

Abstract

Interest in the application of neural networks as tools for decision support has been growing in recent years. A major drawback often associated with neural networks is the difficulty in understanding the knowledge represented by a trained network. This paper describes an approach that can extract symbolic rules from neural networks. We illustrate how the approach successfully extracted rules from a data set collected from a survey of the service sectors in the United Kingdom. The extracted rules were then used to distinguish between organizations using computers from those that do not. The classification scheme based on these rules was used to identify specific segments of a market for promoting adoption of information technology. The extracted rules are not only concise but also outperform discriminant analysis in terms of predictive accuracy.

Keywords: Backpropagation Algorithm, Neural Networks, Symbolic Rules, Technology Adoption

JEL Classification: C45, C38, M15

Suggested Citation

Setiono, Rudy and Thong, James Y.L. and Yap, Chee-Sing, Symbolic Rule Extraction From Neural Networks: An Application to Identifying Organizations Adopting IT (1998). Information & Management, 34, 2 (1998), pp. 91-101., HKUST Business School Research Paper No. 2021-006, Available at SSRN: https://ssrn.com/abstract=3766036

Rudy Setiono

National University of Singapore ( email )

Singapore

James Y.L. Thong (Contact Author)

HKUST Business School ( email )

Clear Water Bay
Kowloon
Hong Kong

HOME PAGE: http://jthong.people.ust.hk/

Chee-Sing Yap

affiliation not provided to SSRN ( email )

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
27
Abstract Views
332
PlumX Metrics