Fast and Accurate Unlexicalized Parsing via Structural Annotations

Maximilian Schlund; Michael Luttenberger; Javier Esparza

Fast and Accurate Unlexicalized Parsing via Structural Annotations

Maximilian Schlund, Michael Luttenberger, Javier Esparza

Informatics 7 - Chair of Theoretical Computer Science

Technical University of Munich

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

We suggest a new annotation scheme for unlexicalized PCFGs that is inspired by formal language theory and only depends on the structure of the parse trees. We evaluate this scheme on the TüBa-D/Z treebank w.r.t. several metrics and show that it improves both parsing accuracy and parsing speed considerably. We also show that our strategy can be fruitfully combined with known ones like parent annotation to achieve accuracies of over 90% labeled F₁ and leaf-ancestor score. Despite increasing the size of the grammar, our annotation allows for parsing more than twice as fast as the PCFG baseline.

Original language	English
Title of host publication	EACL 2014 - 14th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference
Publisher	Association for Computational Linguistics (ACL)
Pages	164-168
Number of pages	5
ISBN (Electronic)	9781937284992
State	Published - 2014
Event	14th Conference of the European Chapter of the Association for Computational Linguistics, EACL 2014 - Gothenburg, Sweden Duration: 26 Apr 2014 → 30 Apr 2014

Publication series

Name	EACL 2014 - 14th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference

Conference

Conference	14th Conference of the European Chapter of the Association for Computational Linguistics, EACL 2014
Country/Territory	Sweden
City	Gothenburg
Period	26/04/14 → 30/04/14

Cite this

Schlund, M., Luttenberger, M., & Esparza, J. (2014). Fast and Accurate Unlexicalized Parsing via Structural Annotations. In EACL 2014 - 14th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference (pp. 164-168). (EACL 2014 - 14th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference). Association for Computational Linguistics (ACL).

Schlund, Maximilian ; Luttenberger, Michael ; Esparza, Javier. / Fast and Accurate Unlexicalized Parsing via Structural Annotations. EACL 2014 - 14th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference. Association for Computational Linguistics (ACL), 2014. pp. 164-168 (EACL 2014 - 14th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference).

@inproceedings{2454518491234e809e778a81f04412fc,

title = "Fast and Accurate Unlexicalized Parsing via Structural Annotations",

abstract = "We suggest a new annotation scheme for unlexicalized PCFGs that is inspired by formal language theory and only depends on the structure of the parse trees. We evaluate this scheme on the T{\"u}Ba-D/Z treebank w.r.t. several metrics and show that it improves both parsing accuracy and parsing speed considerably. We also show that our strategy can be fruitfully combined with known ones like parent annotation to achieve accuracies of over 90% labeled F1 and leaf-ancestor score. Despite increasing the size of the grammar, our annotation allows for parsing more than twice as fast as the PCFG baseline.",

author = "Maximilian Schlund and Michael Luttenberger and Javier Esparza",

note = "Publisher Copyright: {\textcopyright} 2014 Association for Computational Linguistics; 14th Conference of the European Chapter of the Association for Computational Linguistics, EACL 2014 ; Conference date: 26-04-2014 Through 30-04-2014",

year = "2014",

language = "English",

series = "EACL 2014 - 14th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference",

publisher = "Association for Computational Linguistics (ACL)",

pages = "164--168",

booktitle = "EACL 2014 - 14th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference",

}

Schlund, M, Luttenberger, M & Esparza, J 2014, Fast and Accurate Unlexicalized Parsing via Structural Annotations. in EACL 2014 - 14th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference. EACL 2014 - 14th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference, Association for Computational Linguistics (ACL), pp. 164-168, 14th Conference of the European Chapter of the Association for Computational Linguistics, EACL 2014, Gothenburg, Sweden, 26/04/14.

Fast and Accurate Unlexicalized Parsing via Structural Annotations. / Schlund, Maximilian; Luttenberger, Michael ; Esparza, Javier.
EACL 2014 - 14th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference. Association for Computational Linguistics (ACL), 2014. p. 164-168 (EACL 2014 - 14th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Fast and Accurate Unlexicalized Parsing via Structural Annotations

AU - Schlund, Maximilian

AU - Luttenberger, Michael

AU - Esparza, Javier

PY - 2014

Y1 - 2014

N2 - We suggest a new annotation scheme for unlexicalized PCFGs that is inspired by formal language theory and only depends on the structure of the parse trees. We evaluate this scheme on the TüBa-D/Z treebank w.r.t. several metrics and show that it improves both parsing accuracy and parsing speed considerably. We also show that our strategy can be fruitfully combined with known ones like parent annotation to achieve accuracies of over 90% labeled F1 and leaf-ancestor score. Despite increasing the size of the grammar, our annotation allows for parsing more than twice as fast as the PCFG baseline.

AB - We suggest a new annotation scheme for unlexicalized PCFGs that is inspired by formal language theory and only depends on the structure of the parse trees. We evaluate this scheme on the TüBa-D/Z treebank w.r.t. several metrics and show that it improves both parsing accuracy and parsing speed considerably. We also show that our strategy can be fruitfully combined with known ones like parent annotation to achieve accuracies of over 90% labeled F1 and leaf-ancestor score. Despite increasing the size of the grammar, our annotation allows for parsing more than twice as fast as the PCFG baseline.

UR - http://www.scopus.com/inward/record.url?scp=85122025589&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85122025589

T3 - EACL 2014 - 14th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference

SP - 164

EP - 168

BT - EACL 2014 - 14th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference

PB - Association for Computational Linguistics (ACL)

T2 - 14th Conference of the European Chapter of the Association for Computational Linguistics, EACL 2014

Y2 - 26 April 2014 through 30 April 2014

ER -

Schlund M, Luttenberger M , Esparza J. Fast and Accurate Unlexicalized Parsing via Structural Annotations. In EACL 2014 - 14th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference. Association for Computational Linguistics (ACL). 2014. p. 164-168. (EACL 2014 - 14th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference).

Fast and Accurate Unlexicalized Parsing via Structural Annotations

Abstract

Publication series

Conference

Other files and links

Fingerprint

Cite this