JSAI2022

Presentation information

Interactive Session

General Session » Interactive Session

[4Yin2] Interactive session 2

Fri. Jun 17, 2022 12:00 PM - 1:40 PM Room Y (Event Hall)

[4Yin2-53] Synonym Extraction of Failure for Maintenance Knowledge Graph Construction

〇Tomoaki Morioka1, Yasutsugu Morimoto1, Takayuki Uchida1 (1.Hitachi, Ltd.)

Keywords:Knowledge Graph, NLP, Synonym Extraction

The shortage of maintenance worker will become more serious. Therefore, it is required to automate maintenance work by accumulating and utilizing knowledge about equipment to be maintained. The maintenance knowledge graph consists of nodes such as components and failures and their relationships. In this study, we proposed an unsupervised synonym extraction method for failure expression, which is a problem when automatically constructing a knowledge graph from texts such as maintenance reports. One of the existing methods uses the distribution of co-occurrence words. This method may not be accurate in a specialized field. We proposed the method, in which the co-occurrence expression is limited to the information peculiar to the maintenance. As a result, we found that the expressions of the failure, which cannot be estimated as synonymous by the existing method but is synonymous, can be correctly estimated by our method.

Authentication for paper PDF access

A password is required to view paper PDFs. If you are a registered participant, please log on the site from Participant Log In.
You could view the PDF with entering the PDF viewing password bellow.

Password