User Information Extraction for Personalized Dialogue Systems

We propose a method to extract user information in a structured form for personalized dialogue systems. Assuming that user information can be represented as a quadruple <predicate-argument structure, entity, attribute category, topic>, we focus on solving problems in extracting predicate argum...

Full description

Saved in:
Bibliographic Details
Published inTransactions of the Japanese Society for Artificial Intelligence Vol. 31; no. 1; pp. DSF-B_1 - 10
Main Authors Hirano, Toru, Kobayashi, Nozomi, Higashinaka, Ryuichiro, Makino, Toshiro, Matsuo, Yoshihiro
Format Journal Article
LanguageJapanese
Published The Japanese Society for Artificial Intelligence 08.01.2016
Subjects
personalized dialogue systems
predicate-argument structure analysis
user information extraction
Online AccessGet full text

Cover

More Information
Summary:We propose a method to extract user information in a structured form for personalized dialogue systems. Assuming that user information can be represented as a quadruple <predicate-argument structure, entity, attribute category, topic>, we focus on solving problems in extracting predicate argument structures from question-answer pairs in which arguments and predicates are frequently omitted, and in estimating attribute categories related to user behavior which a method using only content words cannot distinguish. Experimental results show that the proposed method significantly outperformed baseline methods and was able to extract user information with 81.2% precision and 58.1% recall.
ISSN:1346-0714
1346-8030
DOI:10.1527/tjsai.DSF-512