WEBIC: a web based business insolvency classifier using neural networks

• Main Authors: Siraj, Fadzilah, Zakaria, Azizi, Ab. Aziz, Azizi, Abas, Zulhazlin
• Format: Conference or Workshop Item
• Language: English
• Published: 2003
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://repo.uum.edu.my/20816/1/AIAI%202003%201%206.pdf; http://repo.uum.edu.my/20816/

Business insolvency is one of the major problems faced by decision makers, especially to detect the early symptom that may contribute to critical business condition.This paper discusses the implementation of neural networks in classifying business insolvency cases in Malaysia. The developed prototype can be accessed remotely via World Wide Web (WWW).For the development purposes, the data was obtained from the Registrar of Business / Companies (ROB/ROC), Kuala Lumpur Stock Exchange and Bank Negara Malaysia (Central Bank of Malaysia).Several experiments were conducted to determine the most suitable parameters for the neural network model.Based on the experimental results, a network with an architecture of 11-6-1 with learning rate 0.1 and momentum term of 0.5. The prototype obtained 90.25% generalization and therefore indicates that the prototype has the potential to be used as a tool for classifying business insolvency.Hence, the prototype provides a basic framework for developing such a classifier