Copenhagen: Tree approximations of semantic parsing problems

Natalie Schluter; Anders Søgaard; Jakob Elming; Dirk Hovy; Barbara Plank; Hector Martinez Alonso; Anders Johanssen; Sigrid Klerke

Copenhagen: Tree approximations of semantic parsing problems

Natalie Schluter
, Anders Søgaard
, Jakob Elming
, Dirk Hovy
, Barbara Plank
, Hector Martinez Alonso
, Anders Johanssen
, Sigrid Klerke

Malmö University

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

Abstract

In this shared task paper for SemEval2014 Task 8, we show that most semantic structures can be approximated by trees through a series of almost bijective graph transformations. We transform input graphs, apply off-the-shelf methods from syntactic parsing on the resulting trees, and retrieve output graphs. Using tree approximations, we obtain good results across three semantic formalisms,
with a 15.9% error reduction over a stateof-the-art semantic role labeling system on development data. Our system came in 3/6 in the shared task closed track.

Original language	English
Title of host publication	Proceedings of the 8th International Workshop on Semantic Evaluation (SemEval 2014)
Publisher	Association for Computational Linguistics
Publication date	2014
Pages	213-217
Publication status	Published - 2014
Externally published	Yes

Keywords

semantic structures
graph transformations
syntactic parsing
semantic formalisms
semantic role labeling

Cite this

@inproceedings{439d01f97034440f8aaba3c808113ee7,

title = "Copenhagen: Tree approximations of semantic parsing problems",

abstract = "In this shared task paper for SemEval2014 Task 8, we show that most semantic structures can be approximated by trees through a series of almost bijective graph transformations. We transform input graphs, apply off-the-shelf methods from syntactic parsing on the resulting trees, and retrieve output graphs. Using tree approximations, we obtain good results across three semantic formalisms, with a 15.9\% error reduction over a stateof-the-art semantic role labeling system on development data. Our system came in 3/6 in the shared task closed track.",

keywords = "semantic structures, graph transformations, syntactic parsing, semantic formalisms, semantic role labeling, semantic structures, graph transformations, syntactic parsing, semantic formalisms, semantic role labeling",

author = "Natalie Schluter and Anders S{\o}gaard and Jakob Elming and Dirk Hovy and Barbara Plank and \{Martinez Alonso\}, Hector and Anders Johanssen and Sigrid Klerke",

year = "2014",

language = "English",

pages = "213--217",

booktitle = "Proceedings of the 8th International Workshop on Semantic Evaluation (SemEval 2014)",

publisher = "Association for Computational Linguistics",

address = "United States",

}

Copenhagen: Tree approximations of semantic parsing problems. / Schluter, Natalie; Søgaard, Anders; Elming, Jakob et al.
Proceedings of the 8th International Workshop on Semantic Evaluation (SemEval 2014). Association for Computational Linguistics, 2014. p. 213-217.

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

TY - GEN

T1 - Copenhagen: Tree approximations of semantic parsing problems

AU - Schluter, Natalie

AU - Søgaard, Anders

AU - Elming, Jakob

AU - Hovy, Dirk

AU - Plank, Barbara

AU - Martinez Alonso, Hector

AU - Johanssen, Anders

AU - Klerke, Sigrid

PY - 2014

Y1 - 2014

N2 - In this shared task paper for SemEval2014 Task 8, we show that most semantic structures can be approximated by trees through a series of almost bijective graph transformations. We transform input graphs, apply off-the-shelf methods from syntactic parsing on the resulting trees, and retrieve output graphs. Using tree approximations, we obtain good results across three semantic formalisms, with a 15.9% error reduction over a stateof-the-art semantic role labeling system on development data. Our system came in 3/6 in the shared task closed track.

AB - In this shared task paper for SemEval2014 Task 8, we show that most semantic structures can be approximated by trees through a series of almost bijective graph transformations. We transform input graphs, apply off-the-shelf methods from syntactic parsing on the resulting trees, and retrieve output graphs. Using tree approximations, we obtain good results across three semantic formalisms, with a 15.9% error reduction over a stateof-the-art semantic role labeling system on development data. Our system came in 3/6 in the shared task closed track.

KW - semantic structures

KW - graph transformations

KW - syntactic parsing

KW - semantic formalisms

KW - semantic role labeling

KW - semantic structures

KW - graph transformations

KW - syntactic parsing

KW - semantic formalisms

KW - semantic role labeling

UR - https://pdfs.semanticscholar.org/6a37/bd5e5ec2cb008ebd1d65cdf3e869f1b38eae.pdf

M3 - Article in proceedings

SP - 213

EP - 217

BT - Proceedings of the 8th International Workshop on Semantic Evaluation (SemEval 2014)

PB - Association for Computational Linguistics

ER -

Copenhagen: Tree approximations of semantic parsing problems

Abstract

Keywords

Fingerprint

Cite this