Development of an adaptive business insolvency classifier prototype (AVICENA) using hybrid intelligent algorithms

• Main Authors: Ab. Aziz, Azizi, Siraj, Fadzilah, Zakaria, Azizi
• Format: Conference or Workshop Item
• Language: English
• Published: 2002
• Subjects: QA76 Computer software
• Online Access: http://repo.uum.edu.my/12329/1/01033085.pdf; http://repo.uum.edu.my/12329/; http://dx.doi.org/10.1109/SCORED.2002.1033085
• Institution: Universiti Utara Malaysia

Confronted by an increasingly competitive environment and chaotic economic conditions, businesses are faced with the need to accept greater risk.Businesses do not become insolvent overnight, rather creditors, investors and the financial community will receive either direct or indirect indications that a company is experiencing financial distress.Thus, this paper analyzed the ability of AVICENA to classify business insolvency performance events.Neural networks (multilayer perceptron-backpropagation) serves as a classifier mechanism while a priori algorithms (auto association rules) support the decision made by the neural networks, in which rules are generated.The conventional model for predicting business performance, the Altman-Z scores model, is used for performance comparison.