Automatic Evaluation of Commonsense Knowledge for Refining Japanese ConceptNet

Seiya Shudo, Rafal Rzepka, Kenji Araki


Abstract
In this paper we present two methods for automatic common sense knowledge evaluation for Japanese entries in ConceptNet ontology. Our proposed methods utilize text-mining approach: one with relation clue words and WordNet synonyms, and one without. Both methods were tested with a blog corpus. The system based on our proposed methods reached relatively high precision score for three relations (MadeOf, UsedFor, AtLocation), which is comparable with previous research using commercial search engines and simpler input. We analyze errors and discuss problems of common sense evaluation, both manual and automatic and propose ideas for further improvements.
Anthology ID:
W16-5413
Volume:
Proceedings of the 12th Workshop on Asian Language Resources (ALR12)
Month:
December
Year:
2016
Address:
Osaka, Japan
Editors:
Koiti Hasida, Kam-Fai Wong, Nicoletta Calzorari, Key-Sun Choi
Venue:
ALR
SIG:
Publisher:
The COLING 2016 Organizing Committee
Note:
Pages:
105–112
Language:
URL:
https://aclanthology.org/W16-5413
DOI:
Bibkey:
Cite (ACL):
Seiya Shudo, Rafal Rzepka, and Kenji Araki. 2016. Automatic Evaluation of Commonsense Knowledge for Refining Japanese ConceptNet. In Proceedings of the 12th Workshop on Asian Language Resources (ALR12), pages 105–112, Osaka, Japan. The COLING 2016 Organizing Committee.
Cite (Informal):
Automatic Evaluation of Commonsense Knowledge for Refining Japanese ConceptNet (Shudo et al., ALR 2016)
Copy Citation:
PDF:
https://aclanthology.org/W16-5413.pdf
Data
NELL