Abstract
In this paper we describe that the hierarchical tag context tree (HTCT) approach improves the accuracy of semantic role labeling on Japanese text. In Japanese language there are functional multiword expressions such as no-tame-ni and yotte that have potential to designate semantic relations between a predicate and its arguments. Since these expressions come to the end part of each argument, the performance of the CRF-based semantic role labeler can be improved by taking into account the last morphemes of each argument as features. We apply our proposed system to the annotated corpus of semantic role labels on a balanced Japanese corpus. The experimental results show that the CRFbased labeler with features extracted by HTCT approach outperforms the normal CRF-based labeler.
| Original language | English |
|---|---|
| Title of host publication | Communications in Computer and Information Science |
| Publisher | Springer Verlag |
| Pages | 21-32 |
| Number of pages | 12 |
| Volume | 593 |
| ISBN (Print) | 9789811005145 |
| DOIs | |
| Publication status | Published - 2016 |
| Event | 14th International Conference of the Pacific Association for Computaitonal Linguistics, PACLING 2015 - Bali, Indonesia Duration: May 19 2015 → May 21 2015 |
Publication series
| Name | Communications in Computer and Information Science |
|---|---|
| Volume | 593 |
| ISSN (Print) | 18650929 |
Other
| Other | 14th International Conference of the Pacific Association for Computaitonal Linguistics, PACLING 2015 |
|---|---|
| Country | Indonesia |
| City | Bali |
| Period | 5/19/15 → 5/21/15 |
Fingerprint
Keywords
- CRFs
- Hierarchical tag context trees
- Semantic role labeling
ASJC Scopus subject areas
- Computer Science(all)
Cite this
Japanese semantic role labeling with hierarchical tag context trees. / Ishihara, Yasuhiro; Takeuchi, Koichi.
Communications in Computer and Information Science. Vol. 593 Springer Verlag, 2016. p. 21-32 (Communications in Computer and Information Science; Vol. 593).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Japanese semantic role labeling with hierarchical tag context trees
AU - Ishihara, Yasuhiro
AU - Takeuchi, Koichi
PY - 2016
Y1 - 2016
N2 - In this paper we describe that the hierarchical tag context tree (HTCT) approach improves the accuracy of semantic role labeling on Japanese text. In Japanese language there are functional multiword expressions such as no-tame-ni and yotte that have potential to designate semantic relations between a predicate and its arguments. Since these expressions come to the end part of each argument, the performance of the CRF-based semantic role labeler can be improved by taking into account the last morphemes of each argument as features. We apply our proposed system to the annotated corpus of semantic role labels on a balanced Japanese corpus. The experimental results show that the CRFbased labeler with features extracted by HTCT approach outperforms the normal CRF-based labeler.
AB - In this paper we describe that the hierarchical tag context tree (HTCT) approach improves the accuracy of semantic role labeling on Japanese text. In Japanese language there are functional multiword expressions such as no-tame-ni and yotte that have potential to designate semantic relations between a predicate and its arguments. Since these expressions come to the end part of each argument, the performance of the CRF-based semantic role labeler can be improved by taking into account the last morphemes of each argument as features. We apply our proposed system to the annotated corpus of semantic role labels on a balanced Japanese corpus. The experimental results show that the CRFbased labeler with features extracted by HTCT approach outperforms the normal CRF-based labeler.
KW - CRFs
KW - Hierarchical tag context trees
KW - Semantic role labeling
UR - http://www.scopus.com/inward/record.url?scp=84961194105&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84961194105&partnerID=8YFLogxK
U2 - 10.1007/978-981-10-0515-2_2
DO - 10.1007/978-981-10-0515-2_2
M3 - Conference contribution
AN - SCOPUS:84961194105
SN - 9789811005145
VL - 593
T3 - Communications in Computer and Information Science
SP - 21
EP - 32
BT - Communications in Computer and Information Science
PB - Springer Verlag
ER -