Natural language processing neural network considering deep cases

Tsukasa Sagara, Masafumi Hagiwara

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

In this paper, we propose a novel neural network considering deep cases. It can learn knowledge from natural language documents and can perform recall and inference. Various techniques of natural language processing using Neural Network have been proposed. However, natural language sentences used in these techniques consist of about a few words, and they cannot handle complicated sentences. In order to solve these problems, the proposed network divides natural language sentences into a sentence layer, a knowledge layer, ten kinds of deep case layers and a dictionary layer. It can learn the relations among sentences and among words by dividing sentences. The advantages of the method are as follows: (1) ability to handle complicated sentences; (2) ability to restructure sentences; (3) usage of the conceptual dictionary, Goi-Taikei, as the long term memory in a brain. Two kinds of experiments were carried out by using goo dictionary and Wikipedia as knowledge sources. Superior performance of the proposed neural network has been confirmed.

Original languageEnglish
Pages (from-to)551-557
Number of pages7
JournalIEEJ Transactions on Electronics, Information and Systems
Volume131
Issue number3
DOIs
Publication statusPublished - 2011

Fingerprint

Glossaries
Processing
Neural networks
Brain
Data storage equipment
Deep neural networks
Experiments

Keywords

  • Deep Case
  • Inference
  • Natural Language Processing
  • Neural Network

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Natural language processing neural network considering deep cases. / Sagara, Tsukasa; Hagiwara, Masafumi.

In: IEEJ Transactions on Electronics, Information and Systems, Vol. 131, No. 3, 2011, p. 551-557.

Research output: Contribution to journalArticle

@article{8874ddb31fca4055bb5538d2d8f58204,
title = "Natural language processing neural network considering deep cases",
abstract = "In this paper, we propose a novel neural network considering deep cases. It can learn knowledge from natural language documents and can perform recall and inference. Various techniques of natural language processing using Neural Network have been proposed. However, natural language sentences used in these techniques consist of about a few words, and they cannot handle complicated sentences. In order to solve these problems, the proposed network divides natural language sentences into a sentence layer, a knowledge layer, ten kinds of deep case layers and a dictionary layer. It can learn the relations among sentences and among words by dividing sentences. The advantages of the method are as follows: (1) ability to handle complicated sentences; (2) ability to restructure sentences; (3) usage of the conceptual dictionary, Goi-Taikei, as the long term memory in a brain. Two kinds of experiments were carried out by using goo dictionary and Wikipedia as knowledge sources. Superior performance of the proposed neural network has been confirmed.",
keywords = "Deep Case, Inference, Natural Language Processing, Neural Network",
author = "Tsukasa Sagara and Masafumi Hagiwara",
year = "2011",
doi = "10.1541/ieejeiss.131.551",
language = "English",
volume = "131",
pages = "551--557",
journal = "IEEJ Transactions on Electronics, Information and Systems",
issn = "0385-4221",
publisher = "The Institute of Electrical Engineers of Japan",
number = "3",

}

TY - JOUR

T1 - Natural language processing neural network considering deep cases

AU - Sagara, Tsukasa

AU - Hagiwara, Masafumi

PY - 2011

Y1 - 2011

N2 - In this paper, we propose a novel neural network considering deep cases. It can learn knowledge from natural language documents and can perform recall and inference. Various techniques of natural language processing using Neural Network have been proposed. However, natural language sentences used in these techniques consist of about a few words, and they cannot handle complicated sentences. In order to solve these problems, the proposed network divides natural language sentences into a sentence layer, a knowledge layer, ten kinds of deep case layers and a dictionary layer. It can learn the relations among sentences and among words by dividing sentences. The advantages of the method are as follows: (1) ability to handle complicated sentences; (2) ability to restructure sentences; (3) usage of the conceptual dictionary, Goi-Taikei, as the long term memory in a brain. Two kinds of experiments were carried out by using goo dictionary and Wikipedia as knowledge sources. Superior performance of the proposed neural network has been confirmed.

AB - In this paper, we propose a novel neural network considering deep cases. It can learn knowledge from natural language documents and can perform recall and inference. Various techniques of natural language processing using Neural Network have been proposed. However, natural language sentences used in these techniques consist of about a few words, and they cannot handle complicated sentences. In order to solve these problems, the proposed network divides natural language sentences into a sentence layer, a knowledge layer, ten kinds of deep case layers and a dictionary layer. It can learn the relations among sentences and among words by dividing sentences. The advantages of the method are as follows: (1) ability to handle complicated sentences; (2) ability to restructure sentences; (3) usage of the conceptual dictionary, Goi-Taikei, as the long term memory in a brain. Two kinds of experiments were carried out by using goo dictionary and Wikipedia as knowledge sources. Superior performance of the proposed neural network has been confirmed.

KW - Deep Case

KW - Inference

KW - Natural Language Processing

KW - Neural Network

UR - http://www.scopus.com/inward/record.url?scp=80052339022&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80052339022&partnerID=8YFLogxK

U2 - 10.1541/ieejeiss.131.551

DO - 10.1541/ieejeiss.131.551

M3 - Article

AN - SCOPUS:80052339022

VL - 131

SP - 551

EP - 557

JO - IEEJ Transactions on Electronics, Information and Systems

JF - IEEJ Transactions on Electronics, Information and Systems

SN - 0385-4221

IS - 3

ER -