Using duo output neural network to solve binary classification problems

Authors:
Pawalai Kraipeerapun;Somkid Amornsamankul
Affiliations:
Ramkhamhaeng University, Faculty of Science, Department of Computer Science, Bangkok, Thailand;Mahidol University, Faculty of Science, Department of Mathematics, Bangkok, Thailand and Centre of Excellence in Mathematics, Bangkok, Thailand
Venue:
ACS'10 Proceedings of the 10th WSEAS international conference on Applied computer science
Year:
2010

Citing 6
Cited 0

Comparative Exudate Classification Using Support Vector Machines and Neural Networks

MICCAI '02 Proceedings of the 5th International Conference on Medical Image Computing and Computer-Assisted Intervention-Part II
One-class document classification via Neural Networks

Neurocomputing
Review: Neural networks and statistical techniques: A review of applications

Expert Systems with Applications: An International Journal
Binary classification using ensemble neural networks and interval neutrosophic sets

Neurocomputing
Applying Duo Output Neural Networks to Solve Single Output Regression Problem

ICONIP '09 Proceedings of the 16th International Conference on Neural Information Processing: Part I
Radial Basis Function Neural Networks Classification for the Recognition of Idiopathic Pulmonary Fibrosis in Microscopic Images

IEEE Transactions on Information Technology in Biomedicine

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes an approach to solve binary classification problems using Duo Output Neural Network (DONN). DONN is a neural network trained to predict a pair of complementary outputs which are the truth and falsity values. In this paper, outputs obtained from two DONNs are aggregated and used to predict the classification result. The first DONN is trained to predict a pair of truth and falsity values. The second DONN is trained to predict a pair of falsity and truth values. The target outputs used to train the second network are organized in reverse order of the first network. The proposed approach has been tested with three benchmarking UCI data sets, which are ionosphere, pima, and liver. It is found that the proposed techniques improve the performance as compared to feedforward backprogation neural network and complementary neural network.