Novel neural network for four-term analogy based on area representation

Kenji Mizoguchi, Masafumi Hagiwara

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we propose a novel neural network for four-term analogy based on area representation. It can deal with four-term analogy such as `teacher:student = doctor:?'. The proposed network is composed of three map layers and an input layer. The area representation method based on Kohonen Feature Map (KFM) is employed in order to represent knowledge, so that similar concepts are mapped in nearer area in the map layer. The proposed mechanism in the map layer can realize the movement of the excited area to the near area. We carried out some computer simulations and confirmed the followings: (1) similar concepts are mapped in the nearer area in the map layer; (2) the excited area moves among similar concepts; (3) the proposed network realizes four-term analogy; (4) the network is robust for the lack of connections.

Original languageEnglish
Title of host publicationProceedings of the International Joint Conference on Neural Networks
PublisherIEEE
Pages1144-1149
Number of pages6
Volume2
Publication statusPublished - 1999
EventInternational Joint Conference on Neural Networks (IJCNN'99) - Washington, DC, USA
Duration: 1999 Jul 101999 Jul 16

Other

OtherInternational Joint Conference on Neural Networks (IJCNN'99)
CityWashington, DC, USA
Period99/7/1099/7/16

Fingerprint

Neural networks
Students
Computer simulation

ASJC Scopus subject areas

  • Software

Cite this

Mizoguchi, K., & Hagiwara, M. (1999). Novel neural network for four-term analogy based on area representation. In Proceedings of the International Joint Conference on Neural Networks (Vol. 2, pp. 1144-1149). IEEE.

Novel neural network for four-term analogy based on area representation. / Mizoguchi, Kenji; Hagiwara, Masafumi.

Proceedings of the International Joint Conference on Neural Networks. Vol. 2 IEEE, 1999. p. 1144-1149.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Mizoguchi, K & Hagiwara, M 1999, Novel neural network for four-term analogy based on area representation. in Proceedings of the International Joint Conference on Neural Networks. vol. 2, IEEE, pp. 1144-1149, International Joint Conference on Neural Networks (IJCNN'99), Washington, DC, USA, 99/7/10.
Mizoguchi K, Hagiwara M. Novel neural network for four-term analogy based on area representation. In Proceedings of the International Joint Conference on Neural Networks. Vol. 2. IEEE. 1999. p. 1144-1149
Mizoguchi, Kenji ; Hagiwara, Masafumi. / Novel neural network for four-term analogy based on area representation. Proceedings of the International Joint Conference on Neural Networks. Vol. 2 IEEE, 1999. pp. 1144-1149
@inproceedings{6ea635b86ff34ee28f6ac733a85f9a0f,
title = "Novel neural network for four-term analogy based on area representation",
abstract = "In this paper, we propose a novel neural network for four-term analogy based on area representation. It can deal with four-term analogy such as `teacher:student = doctor:?'. The proposed network is composed of three map layers and an input layer. The area representation method based on Kohonen Feature Map (KFM) is employed in order to represent knowledge, so that similar concepts are mapped in nearer area in the map layer. The proposed mechanism in the map layer can realize the movement of the excited area to the near area. We carried out some computer simulations and confirmed the followings: (1) similar concepts are mapped in the nearer area in the map layer; (2) the excited area moves among similar concepts; (3) the proposed network realizes four-term analogy; (4) the network is robust for the lack of connections.",
author = "Kenji Mizoguchi and Masafumi Hagiwara",
year = "1999",
language = "English",
volume = "2",
pages = "1144--1149",
booktitle = "Proceedings of the International Joint Conference on Neural Networks",
publisher = "IEEE",

}

TY - GEN

T1 - Novel neural network for four-term analogy based on area representation

AU - Mizoguchi, Kenji

AU - Hagiwara, Masafumi

PY - 1999

Y1 - 1999

N2 - In this paper, we propose a novel neural network for four-term analogy based on area representation. It can deal with four-term analogy such as `teacher:student = doctor:?'. The proposed network is composed of three map layers and an input layer. The area representation method based on Kohonen Feature Map (KFM) is employed in order to represent knowledge, so that similar concepts are mapped in nearer area in the map layer. The proposed mechanism in the map layer can realize the movement of the excited area to the near area. We carried out some computer simulations and confirmed the followings: (1) similar concepts are mapped in the nearer area in the map layer; (2) the excited area moves among similar concepts; (3) the proposed network realizes four-term analogy; (4) the network is robust for the lack of connections.

AB - In this paper, we propose a novel neural network for four-term analogy based on area representation. It can deal with four-term analogy such as `teacher:student = doctor:?'. The proposed network is composed of three map layers and an input layer. The area representation method based on Kohonen Feature Map (KFM) is employed in order to represent knowledge, so that similar concepts are mapped in nearer area in the map layer. The proposed mechanism in the map layer can realize the movement of the excited area to the near area. We carried out some computer simulations and confirmed the followings: (1) similar concepts are mapped in the nearer area in the map layer; (2) the excited area moves among similar concepts; (3) the proposed network realizes four-term analogy; (4) the network is robust for the lack of connections.

UR - http://www.scopus.com/inward/record.url?scp=0033344766&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033344766&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0033344766

VL - 2

SP - 1144

EP - 1149

BT - Proceedings of the International Joint Conference on Neural Networks

PB - IEEE

ER -