Presentation 2007-03-16
A Knowledge Processing Neural Network based on Automatic Concept Hierarchization
Masahiro SAITO, Masafumi HAGIWARA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this report, we propose a knowledge processing neural network which is capable of posteriori and deductively reasoning. The proposed system looks up relations between words in a concept dictionary and co-occurrence dictionary. First, the proposed system divides sentences into the subject words and the other words. Then induces these words are input into two-layer network. Second, the system makes hierarchical structure by using concept dictionary. Third, the system induces general knowledge from individual knowledge. We added a function to respond to questions in natural language with "Yes/No" in order to confirm the validity of proposed system by evaluating the quantity of correct answers.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Neural Network / Natural Language Processing / Automatic Learning / Posteriori Reasoning
Paper # NC2006-211
Date of Issue

Conference Information
Committee NC
Conference Date 2007/3/9(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Knowledge Processing Neural Network based on Automatic Concept Hierarchization
Sub Title (in English)
Keyword(1) Neural Network
Keyword(2) Natural Language Processing
Keyword(3) Automatic Learning
Keyword(4) Posteriori Reasoning
1st Author's Name Masahiro SAITO
1st Author's Affiliation Faculty of Science and Technology, Keio University()
2nd Author's Name Masafumi HAGIWARA
2nd Author's Affiliation Faculty of Science and Technology, Keio University
Date 2007-03-16
Paper # NC2006-211
Volume (vol) vol.106
Number (no) 590
Page pp.pp.-
#Pages 6
Date of Issue