Publication: Applying falsity input to neural networks to solve single output regression problems
3
item.page.dc.date.issued
2011-05-23
item.page.dc.type
item.page.dc.identifier.other
2-s2.0-79956156911
item.page.dc.rights
Mahidol University
item.page.dc.rights.holder
SCOPUS
item.page.dc.identifier.citation
10th WSEAS International Conference on Applied Computer and Applied Computational Science, ACACOS'11. (2011), 72-76
punsarn.item.page.citation
Somkid Amornsamankul, Pawalai Kraipeerapun Applying falsity input to neural networks to solve single output regression problems. 10th WSEAS International Conference on Applied Computer and Applied Computational Science, ACACOS'11. (2011), 72-76. Retrieved from: https://repository.li.mahidol.ac.th/handle/123456789/11803
Research Projects
Organizational Units
Authors
Journal Issue
relationships.isThesisOf
item.page.dc.title
Applying falsity input to neural networks to solve single output regression problems
item.page.dc.contributor.author
item.page.dc.contributor.other
item.page.dc.description.abstract
In this paper, both truth and falsity inputs are used to trained neural networks. Falsity input is the complement of the truth input. Two pairs of neural networks are created. The first pair of neural networks are trained using the truth input whereas the second pair of neural networks are trained using the falsity input. Each pair of neural networks are trained to predict degree of truth and degree of falsity outputs based on the truth and falsity targets, respectively. Two novel techniques are proposed based on these two pairs of neural network. We experiment our proposed techniques to three classical benchmark data sets, which are housing, concrete compressive strength, and computer hardware from the UCI machine learning repository. It is found that our proposed techniques improve the prediction performance when compared to backpropagation neural network and complementary neural networks.