What do we need to know about an unknown word when parsing German
We propose a new type of subword embedding designed to provide more information about unknown compounds, a major source for OOV words in German. We present an extrinsic evaluation where we use the compound embeddings as input to a neural dependency parser and compare the results to the ones obtained...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Chapter/Article Conference Paper |
| Language: | English |
| Published: |
September 2017
|
| In: |
First Workshop on Subword and Character Level Models in NLP - proceedings of the workshop
Year: 2017, Pages: 117-123 |
| DOI: | 10.18653/v1/W17-4117 |
| Online Access: | Verlag, lizenzpflichtig, Volltext: https://doi.org/10.18653/v1/W17-4117 Verlag, lizenzpflichtig, Volltext: https://aclanthology.org/W17-4117 
|
| Author Notes: | Bich-Ngoc Do, Ines Rehbein, Anette Frank |
MARC
| LEADER | 00000naa a2200000 c 4500 | ||
|---|---|---|---|
| 001 | 1870904451 | ||
| 003 | DE-627 | ||
| 005 | 20231122191843.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231122s2017 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.18653/v1/W17-4117 |2 doi | |
| 035 | |a (DE-627)1870904451 | ||
| 035 | |a (DE-599)KXP1870904451 | ||
| 035 | |a (OCoLC)1410406972 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rda | ||
| 041 | |a eng | ||
| 084 | |a 28 |2 sdnb | ||
| 100 | 1 | |a Do, Bich-Ngoc |d 1989- |e VerfasserIn |0 (DE-588)1208354051 |0 (DE-627)1694619745 |4 aut | |
| 245 | 1 | 0 | |a What do we need to know about an unknown word when parsing German |c Bich-Ngoc Do, Ines Rehbein, Anette Frank |
| 264 | 1 | |c September 2017 | |
| 300 | |a 7 | ||
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Gesehen am 22.11.2023 | ||
| 520 | |a We propose a new type of subword embedding designed to provide more information about unknown compounds, a major source for OOV words in German. We present an extrinsic evaluation where we use the compound embeddings as input to a neural dependency parser and compare the results to the ones obtained with other types of embeddings. Our evaluation shows that adding compound embeddings yields a significant improvement of 2% LAS over using word embeddings when no POS information is available. When adding POS embeddings to the input, however, the effect levels out. This suggests that it is not the missing information about the semantics of the unknown words that causes problems for parsing German, but the lack of morphological information for unknown words. To augment our evaluation, we also test the new embeddings in a language modelling task that requires both syntactic and semantic information. | ||
| 700 | 1 | |a Rehbein, Ines |e VerfasserIn |0 (DE-588)1207353833 |0 (DE-627)1693632373 |4 aut | |
| 700 | 1 | |a Frank, Anette |e VerfasserIn |0 (DE-588)1020288108 |0 (DE-627)691172161 |0 (DE-576)36005689X |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |a Workshop on Subword and Character Level Models in NLP (1. : 2017 : Kopenhagen) |t First Workshop on Subword and Character Level Models in NLP - proceedings of the workshop |d Stroudsburg, PA : Association for Computational Linguistics (ACL), 2017 |g (2017), Seite 117-123 |h 1 Online-Ressource (xii, 169 Seiten, 4,14 MB) |w (DE-627)1001283872 |z 9781945626913 |7 nnam |
| 773 | 1 | 8 | |g year:2017 |g pages:117-123 |g extent:7 |a What do we need to know about an unknown word when parsing German |
| 787 | 0 | 8 | |i Forschungsdaten |a Do, Bich-Ngoc, 1989 - |t Head selection parsers and LSTM labelers |d Heidelberg : Universität, 2023 |h 1 Online-Ressource (19 Files) |w (DE-627)1870904370 |
| 856 | 4 | 0 | |u https://doi.org/10.18653/v1/W17-4117 |x Verlag |x Resolving-System |z lizenzpflichtig |3 Volltext |
| 856 | 4 | 0 | |u https://aclanthology.org/W17-4117 |x Verlag |z lizenzpflichtig |3 Volltext |
| 951 | |a AR | ||
| 992 | |a 20231122 | ||
| 993 | |a ConferencePaper | ||
| 994 | |a 2017 | ||
| 998 | |g 1020288108 |a Frank, Anette |m 1020288108:Frank, Anette |d 90000 |d 90500 |e 90000PF1020288108 |e 90500PF1020288108 |k 0/90000/ |k 1/90000/90500/ |p 3 |y j | ||
| 998 | |g 1207353833 |a Rehbein, Ines |m 1207353833:Rehbein, Ines |d 90000 |e 90000PR1207353833 |k 0/90000/ |p 2 | ||
| 998 | |g 1208354051 |a Do, Bich-Ngoc |m 1208354051:Do, Bich-Ngoc |d 90000 |d 90500 |e 90000PD1208354051 |e 90500PD1208354051 |k 0/90000/ |k 1/90000/90500/ |p 1 |x j | ||
| 999 | |a KXP-PPN1870904451 |e 4415876641 | ||
| BIB | |a Y | ||
| JSO | |a {"note":["Gesehen am 22.11.2023"],"type":{"media":"Online-Ressource","bibl":"chapter"},"recId":"1870904451","language":["eng"],"title":[{"title":"What do we need to know about an unknown word when parsing German","title_sort":"What do we need to know about an unknown word when parsing German"}],"person":[{"given":"Bich-Ngoc","family":"Do","role":"aut","display":"Do, Bich-Ngoc","roleDisplay":"VerfasserIn"},{"family":"Rehbein","given":"Ines","display":"Rehbein, Ines","roleDisplay":"VerfasserIn","role":"aut"},{"role":"aut","display":"Frank, Anette","roleDisplay":"VerfasserIn","given":"Anette","family":"Frank"}],"physDesc":[{"extent":"7 S."}],"relHost":[{"part":{"extent":"7","text":"(2017), Seite 117-123","pages":"117-123","year":"2017"},"titleAlt":[{"title":"Proceedings of the First Workshop on Subword and Character Level Models in NLP"}],"recId":"1001283872","language":["eng"],"corporate":[{"display":"Workshop on Subword and Character Level Models in NLP (1., 2017, Kopenhagen)","roleDisplay":"VerfasserIn","role":"aut"},{"role":"isb","roleDisplay":"Herausgebendes Organ","display":"Association for Computational Linguistics"}],"type":{"bibl":"book","media":"Online-Ressource"},"disp":"Workshop on Subword and Character Level Models in NLP (1. : 2017 : Kopenhagen)First Workshop on Subword and Character Level Models in NLP - proceedings of the workshop","note":["\"Editors: Manaal Faruqui, Hinrich Schuetze, Isabel Trancoso, Yadollah Yaghoobzadeh\" - Startseite der Ressource","Literaturangaben"],"title":[{"title_sort":"First Workshop on Subword and Character Level Models in NLP - proceedings of the workshop","title":"First Workshop on Subword and Character Level Models in NLP - proceedings of the workshop","subtitle":"September 7, 2017, Copenhagen, Denmark : EMNLP 2017"}],"person":[{"role":"edt","roleDisplay":"HerausgeberIn","display":"Faruqui, Manaal","given":"Manaal","family":"Faruqui"},{"role":"edt","roleDisplay":"HerausgeberIn","display":"Schuetze, Hinrich","given":"Hinrich","family":"Schuetze"},{"family":"Trancoso","given":"Isabel","display":"Trancoso, Isabel","roleDisplay":"HerausgeberIn","role":"edt"},{"given":"Yadollah","family":"Yaghoobzadeh","role":"edt","display":"Yaghoobzadeh, Yadollah","roleDisplay":"HerausgeberIn"}],"physDesc":[{"noteIll":"Illustrationen","extent":"1 Online-Ressource (xii, 169 Seiten, 4,14 MB)"}],"id":{"isbn":["9781945626913"],"eki":["1001283872"]},"origin":[{"dateIssuedDisp":"[2017]","publisher":"Association for Computational Linguistics (ACL)","dateIssuedKey":"2017","publisherPlace":"Stroudsburg, PA"}]}],"origin":[{"dateIssuedKey":"2017","dateIssuedDisp":"September 2017"}],"id":{"doi":["10.18653/v1/W17-4117"],"eki":["1870904451"]},"name":{"displayForm":["Bich-Ngoc Do, Ines Rehbein, Anette Frank"]}} | ||
| SRT | |a DOBICHNGOCWHATDOWENE2017 | ||