Please use this identifier to cite or link to this item:
|Title:||Applying Duo Output Neural Networks to Solve Single Output Regression Problem|
|Keywords:||Complementary Neural Network;organized;polynomial|
|Publisher:||International Conference on Neural Information Processing|
|Abstract:||This paper proposes a novel approach to solve a single output regression problem using duo output neural network. A pair of duo output neural networks is created. The first neural network is trained to provide two outputs which are the truth and the falsity values. The second neural network is also trained to provide two outputs; however, the sequence of the outputs is organized in reverse order of the first one. Therefore, the two outputs of this neural network is the falsity and the truth values. All the truth and the non-falsity values obtained from both neural networks are then averaged to give the final output. We experiment our proposed approach to the classical benchmark problems which are housing, concrete compressive strength, and computer hardware data sets from the UCI machine learning repository. It is found that the proposed approach provides better performance when compared to the complementary neural networks, backpropagation neural networks, and support vector regression with linear, polynomial, and radial basis function kernels.|
|Appears in Collections:||Mathematics: International Proceedings|
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.