Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
RELATIONSHIP ESTIMATION MODEL LEARNING DEVICE, METHOD, AND PROGRAM
Document Type and Number:
WIPO Patent Application WO/2019/160096
Kind Code:
A1
Abstract:
According to the present invention, the relationship between phrases can be accurately estimated while the cost of creating learning data is required. A learning data generation unit 62 extracts, on the basis of the dependency analysis result of an inputted text, paired phrases having a dependency relationship with a clause including a predetermined connection expression indicating the relationship between the phrases, and generates a triple set consisting of the extracted paired phrases and a relationship label representing the connection expression or representing the relationship indicated by the connection expression. A learning unit 63 learns a relationship estimation model for estimating the relationship between the phrases on the basis of the triple set generated by the learning data generation unit.

Inventors:
SAITO ITSUMI (JP)
NISHIDA KYOSUKE (JP)
TOMITA JUNJI (JP)
ASANO HISAKO (JP)
Application Number:
PCT/JP2019/005620
Publication Date:
August 22, 2019
Filing Date:
February 15, 2019
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
NIPPON TELEGRAPH & TELEPHONE (JP)
International Classes:
G06F16/00; G06F17/27; G06N5/02; G06N20/00
Foreign References:
JP2016170636A2016-09-23
Other References:
INUI TAKASHI ET AL: "Acquiring causal knowledge from text using the connective marker tame", IPSJ JOURNAL, vol. 45, no. 3, 15 March 2004 (2004-03-15), pages 919 - 933, XP055634904
SAITO ITSUMI ET AL: "Simultaneous Learning of Complement and Generation of Phrase Knowledge", PROCEEDING OF THE 24TH ANNUAL MEETING OF THE ASSOCIATION FOR NATURAL LANGUAGE PROCESSING, vol. 9, no. 5, 5 March 2018 (2018-03-05), pages 951 - 954, XP055634906, Retrieved from the Internet
Attorney, Agent or Firm:
TAIYO, NAKAJIMA & KATO (JP)
Download PDF: