The Seemingly (Un)systematic Linking Element in Danish
Publikation: Bidrag til tidsskrift › Konferenceartikel › Forskning › fagfællebedømt
Standard
The Seemingly (Un)systematic Linking Element in Danish. / Boldsen, Sidsel; Aguirrezabal Zabaleta, Manex.
I: NEALT (Northern European Association of Language Technology) Monograph Series, Bind 42, 10.2019, s. 376-380.Publikation: Bidrag til tidsskrift › Konferenceartikel › Forskning › fagfællebedømt
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - The Seemingly (Un)systematic Linking Element in Danish
AU - Boldsen, Sidsel
AU - Aguirrezabal Zabaleta, Manex
PY - 2019/10
Y1 - 2019/10
N2 - The use of a linking element between compound members is a common phenomenon in Germanic languages. Still, the exact use and conditioning of such elements is a disputed topic in linguistics. In this paper we address the issue of predicting the use of linking elements in Danish. Following previous research that shows how the choice of linking element might be conditioned by phonology, we frame the problem as a language modeling task: Considering the linking elements -s/-∅ the problem becomes predicting what is most probable to encounter next, a syllable boundary or the joining element, s. We show that training a language model on this task reaches an accuracy of 94 %, and in the case of an unsupervised model, the accuracy reaches 80 %.
AB - The use of a linking element between compound members is a common phenomenon in Germanic languages. Still, the exact use and conditioning of such elements is a disputed topic in linguistics. In this paper we address the issue of predicting the use of linking elements in Danish. Following previous research that shows how the choice of linking element might be conditioned by phonology, we frame the problem as a language modeling task: Considering the linking elements -s/-∅ the problem becomes predicting what is most probable to encounter next, a syllable boundary or the joining element, s. We show that training a language model on this task reaches an accuracy of 94 %, and in the case of an unsupervised model, the accuracy reaches 80 %.
M3 - Conference article
VL - 42
SP - 376
EP - 380
JO - NEALT (Northern European Association of Language Technology) Monograph Series
JF - NEALT (Northern European Association of Language Technology) Monograph Series
SN - 1736-6291
ER -
ID: 238738445