Title:
EXTRACTION METHOD, EXTRACTION DEVICE, AND EXTRACTION PROGRAM
Document Type and Number:
WIPO Patent Application WO/2020/009208
Kind Code:
A1
Abstract:
A paragraph dividing unit (15a) divides a document having markup information applied thereto that indicates a vertex of a graph structure, dividing same into paragraphs. A vertex extraction unit (15b) extracts a vertex for each paragraph. An edge extraction unit (15c) extracts edges between extracted vertices, for each paragraph.
A storage unit (14) stores a schema (14a) that specifies edges between the attributes of vertices in the graph structure. The paragraph dividing unit (15a) divides into paragraphs a document having marking information applied thereto that indicates vertices in the graph structure and attributes of the vertices. The vertex extraction unit (15b) extracts the vertices and the attributes of the vertices, for each paragraph. The edge extraction unit (15c) can extract an edge specified in the schema (14a), for each paragraph. The vertex joining unit (15d) can join extracted vertices that are duplicated between paragraphs.
Inventors:
YAMASAKI MAYO (JP)
KAWABE HIDEKI (JP)
KAWABE HIDEKI (JP)
Application Number:
PCT/JP2019/026736
Publication Date:
January 09, 2020
Filing Date:
July 04, 2019
Export Citation:
Assignee:
NIPPON TELEGRAPH & TELEPHONE (JP)
International Classes:
G06F16/36
Foreign References:
| US20150310096A1 | 2015-10-29 |
Other References:
YAMASAKI, MAYO ET AL.: "Automation Extraction of Threat and Vulnerability Knowledge from Documents", SYMPOSIUM SERIES OF INFORMATION PROCESSING SOCIETY OF JAPAN, vol. 2017, no. 2, 16 October 2017 (2017-10-16), pages 393 - 398
Attorney, Agent or Firm:
SAKAI INTERNATIONAL PATENT OFFICE (JP)
Download PDF: