A knowledge processing neural network based on automatic concept hierarchization

Masahiro Saito, Masafumi Hagiwara

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

Abstract

In this paper, we propose a knowledge processing neural network which is capable of inductive and deductive inference. The proposed network looks up relations between words in a concept dictionary and co-occurrence dictionary. First, the proposed network divides sentences into the subject words and the other words. Then these words are input into two-layer network. Second, hierarchical structure is composed using concept dictionary. Third, the network induces general knowledge from individual knowledge. We added a function to respond to questions in natural language with "Yes/No/I don't know" in order to confirm the validity of proposed network by evaluating the quantity of correct answers.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages539-548
Number of pages10
Volume4985 LNCS
EditionPART 2
DOIs
Publication statusPublished - 2008
Event14th International Conference on Neural Information Processing, ICONIP 2007 - Kitakyushu, Japan
Duration: 2007 Nov 132007 Nov 16

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 2
Volume4985 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other14th International Conference on Neural Information Processing, ICONIP 2007
CountryJapan
CityKitakyushu
Period07/11/1307/11/16

Fingerprint

Glossaries
Neural Networks
Neural networks
Processing
Network layers
Hierarchical Structure
Natural Language
Divides
Knowledge
Concepts
Dictionary

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Saito, M., & Hagiwara, M. (2008). A knowledge processing neural network based on automatic concept hierarchization. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (PART 2 ed., Vol. 4985 LNCS, pp. 539-548). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4985 LNCS, No. PART 2). https://doi.org/10.1007/978-3-540-69162-4_56

A knowledge processing neural network based on automatic concept hierarchization. / Saito, Masahiro; Hagiwara, Masafumi.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 4985 LNCS PART 2. ed. 2008. p. 539-548 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4985 LNCS, No. PART 2).

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

Saito, M & Hagiwara, M 2008, A knowledge processing neural network based on automatic concept hierarchization. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 2 edn, vol. 4985 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 2, vol. 4985 LNCS, pp. 539-548, 14th International Conference on Neural Information Processing, ICONIP 2007, Kitakyushu, Japan, 07/11/13. https://doi.org/10.1007/978-3-540-69162-4_56
Saito M, Hagiwara M. A knowledge processing neural network based on automatic concept hierarchization. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 2 ed. Vol. 4985 LNCS. 2008. p. 539-548. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2). https://doi.org/10.1007/978-3-540-69162-4_56
Saito, Masahiro ; Hagiwara, Masafumi. / A knowledge processing neural network based on automatic concept hierarchization. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 4985 LNCS PART 2. ed. 2008. pp. 539-548 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2).
@inproceedings{0d1e59f86cf64b3d88edadfcb21d470e,
title = "A knowledge processing neural network based on automatic concept hierarchization",
abstract = "In this paper, we propose a knowledge processing neural network which is capable of inductive and deductive inference. The proposed network looks up relations between words in a concept dictionary and co-occurrence dictionary. First, the proposed network divides sentences into the subject words and the other words. Then these words are input into two-layer network. Second, hierarchical structure is composed using concept dictionary. Third, the network induces general knowledge from individual knowledge. We added a function to respond to questions in natural language with {"}Yes/No/I don't know{"} in order to confirm the validity of proposed network by evaluating the quantity of correct answers.",
author = "Masahiro Saito and Masafumi Hagiwara",
year = "2008",
doi = "10.1007/978-3-540-69162-4_56",
language = "English",
isbn = "3540691596",
volume = "4985 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
number = "PART 2",
pages = "539--548",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
edition = "PART 2",

}

TY - GEN

T1 - A knowledge processing neural network based on automatic concept hierarchization

AU - Saito, Masahiro

AU - Hagiwara, Masafumi

PY - 2008

Y1 - 2008

N2 - In this paper, we propose a knowledge processing neural network which is capable of inductive and deductive inference. The proposed network looks up relations between words in a concept dictionary and co-occurrence dictionary. First, the proposed network divides sentences into the subject words and the other words. Then these words are input into two-layer network. Second, hierarchical structure is composed using concept dictionary. Third, the network induces general knowledge from individual knowledge. We added a function to respond to questions in natural language with "Yes/No/I don't know" in order to confirm the validity of proposed network by evaluating the quantity of correct answers.

AB - In this paper, we propose a knowledge processing neural network which is capable of inductive and deductive inference. The proposed network looks up relations between words in a concept dictionary and co-occurrence dictionary. First, the proposed network divides sentences into the subject words and the other words. Then these words are input into two-layer network. Second, hierarchical structure is composed using concept dictionary. Third, the network induces general knowledge from individual knowledge. We added a function to respond to questions in natural language with "Yes/No/I don't know" in order to confirm the validity of proposed network by evaluating the quantity of correct answers.

UR - http://www.scopus.com/inward/record.url?scp=54149104118&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=54149104118&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-69162-4_56

DO - 10.1007/978-3-540-69162-4_56

M3 - Conference contribution

SN - 3540691596

SN - 9783540691594

VL - 4985 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 539

EP - 548

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -