Unsupervised Discourse Constituency Parsing Using Viterbi EM

被引：10

作者：

Nishida, Noriki ^{[1
]}

Nakayama, Hideki ^{[1
]}

机构：

[1] Univ Tokyo, Grad Sch Informat Sci & Technol, Tokyo, Japan

来源：

TRANSACTIONS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS | 2020年 / 8卷 / 08期

关键词：

57;

D O I：

10.1162/tacl_a_00312

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

In this paper, we introduce an unsupervised discourse constituency parsing algorithm. We use Viterbi EM with a margin- based criterion to train a span-based discourse parser in an unsupervised manner. We also propose initialization methods for Viterbi training of discourse constituents based on our prior knowledge of text structures. Experimental results demonstrate that our unsupervised parser achieves comparable or even superior performance to fully supervised parsers. We also investigate discourse constituents that are learned by our method.

引用

页码：215 / 230

页数：16

共 50 条

[21] Order-sensitive Neural Constituency Parsing
Wang, Zhicheng
Shi, Tianyu
Xiao, Liyin
Liu, Cong
2022 IEEE 34TH INTERNATIONAL CONFERENCE ON TOOLS WITH ARTIFICIAL INTELLIGENCE, ICTAI, 2022, : 282 - 287
[22] Constituency Parsing by Cross-Lingual Delexicalization
Kaing, Hour
Ding, Chenchen
Utiyama, Masao
Sumita, Eiichiro
Sudoh, Katsuhito
Nakamura, Satoshi
IEEE ACCESS, 2021, 9 : 141571 - 141578
[23] Challenges to Open-Domain Constituency Parsing
Yang, Sen
Cui, Leyang
Ning, Ruoxi
Wu, Di
Zhang, Yue
FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2022), 2022, : 112 - 127
[24] Fast and Accurate Neural CRF Constituency Parsing
Zhang, Yu
Zhou, Houquan
Li, Zhenghua
PROCEEDINGS OF THE TWENTY-NINTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2020, : 4046 - 4053
[25] Constituency Parsing with a Self-Attentive Encoder
Kitaev, Nikita
Klein, Dan
PROCEEDINGS OF THE 56TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL), VOL 1, 2018, : 2676 - 2686
[26] Improving Sequence-to-Sequence Constituency Parsing
Liu, Lemao
Zhu, Muhua
Shi, Shuming
THIRTY-SECOND AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE / THIRTIETH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE / EIGHTH AAAI SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2018, : 4873 - 4880
[27] A* parsing: Fast extract viterbi parse selection
Klein, D
Manning, CD
HLT-NAACL 2003: HUMAN LANGUAGE TECHNOLOGY CONFERENCE OF THE NORTH AMERICAN CHAPTER OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, PROCEEDINGS OF THE MAIN CONFERENCE, 2003, : 119 - 126
[28] Efficient error-correcting viterbi parsing
Amengual, JC
Vidal, E
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1998, 20 (10) : 1109 - 1116
[29] Assessing the Use of Prosody in Constituency Parsing of Imperfect Transcripts
Tran, Trang
Ostendorf, Mari
INTERSPEECH 2021, 2021, : 2626 - 2630
[30] Structured Tree Alignment for Evaluation of (Speech) Constituency Parsing
Shi, Freda
Gimpel, Kevin
Livescu, Karen
PROCEEDINGS OF THE 62ND ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, VOL 1: LONG PAPERS, 2024, : 12320 - 12332

← 1 2 3 4 5 →