Logical rule extraction from data by maximum neural networks

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

3 Citations (Scopus)

Abstract

In this paper, a new neural computing method to extract logical rules from the training data sets is proposed. Maximum neural networks are used to train the weight and the threshold of the multi-layered (feedforward) neural network (MLNN). The threshold and the weights of the MLNN are trained to be a logical function (AND/OR) with the multiple input. The maximum neural network constructs the logical function on the MLNN so that it is not necessary to extract rules from the trained MLNN. The proposed method was experimented for the classification problem, Monk's problem 1. Experimental results showed that the proposed method learned the correct rule in more than 40% success rate.

Original languageEnglish
Title of host publicationProceedings of the 2nd International Conference on Intelligent Processing and Manufacturing of Materials, IPMM 1999
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages723-728
Number of pages6
Volume2
ISBN (Electronic)0780354893, 9780780354890
DOIs
Publication statusPublished - 1999 Jan 1
Event2nd International Conference on Intelligent Processing and Manufacturing of Materials, IPMM 1999 - Honolulu, United States
Duration: 1999 Jul 101999 Jul 15

Other

Other2nd International Conference on Intelligent Processing and Manufacturing of Materials, IPMM 1999
CountryUnited States
CityHonolulu
Period99/7/1099/7/15

Fingerprint

Neural networks
Feedforward neural networks

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering
  • Materials Science (miscellaneous)

Cite this

Saito, T., & Takefuji, Y. (1999). Logical rule extraction from data by maximum neural networks. In Proceedings of the 2nd International Conference on Intelligent Processing and Manufacturing of Materials, IPMM 1999 (Vol. 2, pp. 723-728). [791477] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IPMM.1999.791477

Logical rule extraction from data by maximum neural networks. / Saito, T.; Takefuji, Yoshiyasu.

Proceedings of the 2nd International Conference on Intelligent Processing and Manufacturing of Materials, IPMM 1999. Vol. 2 Institute of Electrical and Electronics Engineers Inc., 1999. p. 723-728 791477.

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

Saito, T & Takefuji, Y 1999, Logical rule extraction from data by maximum neural networks. in Proceedings of the 2nd International Conference on Intelligent Processing and Manufacturing of Materials, IPMM 1999. vol. 2, 791477, Institute of Electrical and Electronics Engineers Inc., pp. 723-728, 2nd International Conference on Intelligent Processing and Manufacturing of Materials, IPMM 1999, Honolulu, United States, 99/7/10. https://doi.org/10.1109/IPMM.1999.791477
Saito T, Takefuji Y. Logical rule extraction from data by maximum neural networks. In Proceedings of the 2nd International Conference on Intelligent Processing and Manufacturing of Materials, IPMM 1999. Vol. 2. Institute of Electrical and Electronics Engineers Inc. 1999. p. 723-728. 791477 https://doi.org/10.1109/IPMM.1999.791477
Saito, T. ; Takefuji, Yoshiyasu. / Logical rule extraction from data by maximum neural networks. Proceedings of the 2nd International Conference on Intelligent Processing and Manufacturing of Materials, IPMM 1999. Vol. 2 Institute of Electrical and Electronics Engineers Inc., 1999. pp. 723-728
@inproceedings{c65c348659fe4c9fa9664d3e808ad30e,
title = "Logical rule extraction from data by maximum neural networks",
abstract = "In this paper, a new neural computing method to extract logical rules from the training data sets is proposed. Maximum neural networks are used to train the weight and the threshold of the multi-layered (feedforward) neural network (MLNN). The threshold and the weights of the MLNN are trained to be a logical function (AND/OR) with the multiple input. The maximum neural network constructs the logical function on the MLNN so that it is not necessary to extract rules from the trained MLNN. The proposed method was experimented for the classification problem, Monk's problem 1. Experimental results showed that the proposed method learned the correct rule in more than 40{\%} success rate.",
author = "T. Saito and Yoshiyasu Takefuji",
year = "1999",
month = "1",
day = "1",
doi = "10.1109/IPMM.1999.791477",
language = "English",
volume = "2",
pages = "723--728",
booktitle = "Proceedings of the 2nd International Conference on Intelligent Processing and Manufacturing of Materials, IPMM 1999",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Logical rule extraction from data by maximum neural networks

AU - Saito, T.

AU - Takefuji, Yoshiyasu

PY - 1999/1/1

Y1 - 1999/1/1

N2 - In this paper, a new neural computing method to extract logical rules from the training data sets is proposed. Maximum neural networks are used to train the weight and the threshold of the multi-layered (feedforward) neural network (MLNN). The threshold and the weights of the MLNN are trained to be a logical function (AND/OR) with the multiple input. The maximum neural network constructs the logical function on the MLNN so that it is not necessary to extract rules from the trained MLNN. The proposed method was experimented for the classification problem, Monk's problem 1. Experimental results showed that the proposed method learned the correct rule in more than 40% success rate.

AB - In this paper, a new neural computing method to extract logical rules from the training data sets is proposed. Maximum neural networks are used to train the weight and the threshold of the multi-layered (feedforward) neural network (MLNN). The threshold and the weights of the MLNN are trained to be a logical function (AND/OR) with the multiple input. The maximum neural network constructs the logical function on the MLNN so that it is not necessary to extract rules from the trained MLNN. The proposed method was experimented for the classification problem, Monk's problem 1. Experimental results showed that the proposed method learned the correct rule in more than 40% success rate.

UR - http://www.scopus.com/inward/record.url?scp=30344442405&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=30344442405&partnerID=8YFLogxK

U2 - 10.1109/IPMM.1999.791477

DO - 10.1109/IPMM.1999.791477

M3 - Conference contribution

AN - SCOPUS:30344442405

VL - 2

SP - 723

EP - 728

BT - Proceedings of the 2nd International Conference on Intelligent Processing and Manufacturing of Materials, IPMM 1999

PB - Institute of Electrical and Electronics Engineers Inc.

ER -