Multi-Version Systems of Neural networks for Predicting the Risk of Osteoporosis

Wenjia Wang, Derek Partridge and Sarah Rae

This paper describes the methodology of developing multi-version systems using neural networks in the hope of improving their performance and reliability. However, a system implemented by simply combining N neural nets may not necessarily deliver a better result than the individual versions alone. A critical factor to success is the diversity among these versions, which is high probability that the system will avoid coincident failures and therefore exhibit increased reliability. Coincident-failure diversity (CFD) is described as a specific measure of the diversity quantitatively. The approach of Multi-Net System(MNS) has been applied to predict the risk of osteoporosis for female patients. The performance of the MNS showed with ROC curves are considerably better than that of the individual nets in the system and also Logistic regression.

This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.