Supersense tagging for Danish

Hector Martinez Alonso; Anders Trærup Johannsen; Sussi Olsen; Sanni Nimb; Nicolai Sørensen; Anna Braasch; Anders Søgaard; Bolette Sandford Pedersen

Supersense tagging for Danish

Hector Martinez Alonso, Anders Trærup Johannsen, Sussi Olsen, Sanni Nimb, Nicolai Sørensen, Anna Braasch, Anders Søgaard, Bolette Sandford Pedersen

Abstract

We describe the creation of a new Danish resource for automated coarse-grained word sense disambiguation of running text (supersense tagging, SST). Based on corpus evidence we expand the sense inventory to incorporate new lexical classes. We add tags for verbal satellites like collocates, particles and reflexive pronouns, to give account for the satellite-framing properties of Danish. Finally, we evaluate the quality of our expanded sense inventory in terms of variation in F1 on a stateof- the- art SST system. The SST systems uses type constraints and achieves performance just under the upper bound of interannotator agreement. The initial release is a 1,500-sentence corpus covering six genres, made available under an open-source license.

Original language	English
Title of host publication	Proceedings of the 20th Nordic Conference of Computational Linguistics NODALIDA 2015
Number of pages	8
Volume	109
Publisher	Linköping University Electronic Press
Publication date	2015
ISBN (Print)	978-91-7519-098-3
Publication status	Published - 2015
Event	NODALIDA 2015: Nordic Conference on Computational Linguistics - Vilnius, Lithuania Duration: 11 May 2015 → 13 May 2015 Conference number: 20

Conference

Conference	NODALIDA 2015
Number	20
Country/Territory	Lithuania
City	Vilnius
Period	11/05/2015 → 13/05/2015

Series	NEALT (Northern European Association of Language Technology) Proceedings Series
Volume	23
ISSN	1736-6305

Cite this

Martinez Alonso, H., Johannsen, A. T., Olsen, S., Nimb, S., Sørensen, N., Braasch, A., Søgaard, A., & Pedersen, B. S. (2015). Supersense tagging for Danish. In Proceedings of the 20th Nordic Conference of Computational Linguistics NODALIDA 2015 (Vol. 109). Linköping University Electronic Press. NEALT (Northern European Association of Language Technology) Proceedings Series Vol. 23

Supersense tagging for Danish. / Martinez Alonso, Hector; Johannsen, Anders Trærup; Olsen, Sussi et al.

Proceedings of the 20th Nordic Conference of Computational Linguistics NODALIDA 2015. Vol. 109 Linköping University Electronic Press, 2015. (NEALT (Northern European Association of Language Technology) Proceedings Series, Vol. 23).

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

Martinez Alonso, H, Johannsen, AT, Olsen, S, Nimb, S, Sørensen, N, Braasch, A, Søgaard, A & Pedersen, BS 2015, Supersense tagging for Danish. in Proceedings of the 20th Nordic Conference of Computational Linguistics NODALIDA 2015. vol. 109, Linköping University Electronic Press, NEALT (Northern European Association of Language Technology) Proceedings Series, vol. 23, NODALIDA 2015, Vilnius, Lithuania, 11/05/2015.

@inproceedings{05f8402f047e4a1c9245dcde74a0b17e,

title = "Supersense tagging for Danish",

abstract = "We describe the creation of a new Danish resource for automated coarse-grained word sense disambiguation of running text (supersense tagging, SST). Based on corpus evidence we expand the sense inventory to incorporate new lexical classes. We add tags for verbal satellites like collocates, particles and reflexive pronouns, to give account for the satellite-framing properties of Danish. Finally, we evaluate the quality of our expanded sense inventory in terms of variation in F1 on a stateof- the- art SST system. The SST systems uses type constraints and achieves performance just under the upper bound of interannotator agreement. The initial release is a 1,500-sentence corpus covering six genres, made available under an open-source license.",

author = "{Martinez Alonso}, Hector and Johannsen, {Anders Tr{\ae}rup} and Sussi Olsen and Sanni Nimb and Nicolai S{\o}rensen and Anna Braasch and Anders S{\o}gaard and Pedersen, {Bolette Sandford}",

note = "Der er ikke overensstemmelse mellem det ISSN-nr der st{\aa}r p{\aa} proceedings og det der findes i databasen. ; NODALIDA 2015 : Nordic Conference on Computational Linguistics ; Conference date: 11-05-2015 Through 13-05-2015",

year = "2015",

language = "English",

isbn = "978-91-7519-098-3",

volume = "109",

series = "NEALT (Northern European Association of Language Technology) Proceedings Series",

booktitle = "Proceedings of the 20th Nordic Conference of Computational Linguistics NODALIDA 2015",

publisher = "Link{\"o}ping University Electronic Press",

}

TY - GEN

T1 - Supersense tagging for Danish

AU - Martinez Alonso, Hector

AU - Johannsen, Anders Trærup

AU - Olsen, Sussi

AU - Nimb, Sanni

AU - Sørensen, Nicolai

AU - Braasch, Anna

AU - Søgaard, Anders

AU - Pedersen, Bolette Sandford

N1 - Conference code: 20

PY - 2015

Y1 - 2015

N2 - We describe the creation of a new Danish resource for automated coarse-grained word sense disambiguation of running text (supersense tagging, SST). Based on corpus evidence we expand the sense inventory to incorporate new lexical classes. We add tags for verbal satellites like collocates, particles and reflexive pronouns, to give account for the satellite-framing properties of Danish. Finally, we evaluate the quality of our expanded sense inventory in terms of variation in F1 on a stateof- the- art SST system. The SST systems uses type constraints and achieves performance just under the upper bound of interannotator agreement. The initial release is a 1,500-sentence corpus covering six genres, made available under an open-source license.

AB - We describe the creation of a new Danish resource for automated coarse-grained word sense disambiguation of running text (supersense tagging, SST). Based on corpus evidence we expand the sense inventory to incorporate new lexical classes. We add tags for verbal satellites like collocates, particles and reflexive pronouns, to give account for the satellite-framing properties of Danish. Finally, we evaluate the quality of our expanded sense inventory in terms of variation in F1 on a stateof- the- art SST system. The SST systems uses type constraints and achieves performance just under the upper bound of interannotator agreement. The initial release is a 1,500-sentence corpus covering six genres, made available under an open-source license.

M3 - Article in proceedings

SN - 978-91-7519-098-3

VL - 109

T3 - NEALT (Northern European Association of Language Technology) Proceedings Series

BT - Proceedings of the 20th Nordic Conference of Computational Linguistics NODALIDA 2015

PB - Linköping University Electronic Press

T2 - NODALIDA 2015

Y2 - 11 May 2015 through 13 May 2015

ER -

Supersense tagging for Danish

Abstract

Conference

Fingerprint

Cite this