SIGTYP 2020 Shared Task: Prediction of Typological Features
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Standard
SIGTYP 2020 Shared Task: Prediction of Typological Features. / Bjerva, Johannes; Salesky, Elizabeth; Mielke, Sabrina J.; Chaudhary, Aditi; Giuseppe, Celano; Ponti, Edoardo Maria; Vylomova, Ekaterina; Cotterell, Ryan; Augenstein, Isabelle.
Proceedings of the Second Workshop on Computational Research in Linguistic Typology. Association for Computational Linguistics, 2020. p. 1-11.Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - SIGTYP 2020 Shared Task: Prediction of Typological Features
AU - Bjerva, Johannes
AU - Salesky, Elizabeth
AU - Mielke, Sabrina J.
AU - Chaudhary, Aditi
AU - Giuseppe, Celano
AU - Ponti, Edoardo Maria
AU - Vylomova, Ekaterina
AU - Cotterell, Ryan
AU - Augenstein, Isabelle
PY - 2020
Y1 - 2020
N2 - Typological knowledge bases (KBs) such as WALS (Dryer and Haspelmath, 2013) contain information about linguistic properties of the world’s languages. They have been shown to be useful for downstream applications, including cross-lingual transfer learning and linguistic probing. A major drawback hampering broader adoption of typological KBs is that they are sparsely populated, in the sense that most languages only have annotations for some features, and skewed, in that few features have wide coverage. As typological features often correlate with one another, it is possible to predict them and thus automatically populate typological KBs, which is also the focus of this shared task. Overall, the task attracted 8 submissions from 5 teams, out of which the most successful methods make use of such feature correlations. However, our error analysis reveals that even the strongest submitted systems struggle with predicting feature values for languages where few features are known.
AB - Typological knowledge bases (KBs) such as WALS (Dryer and Haspelmath, 2013) contain information about linguistic properties of the world’s languages. They have been shown to be useful for downstream applications, including cross-lingual transfer learning and linguistic probing. A major drawback hampering broader adoption of typological KBs is that they are sparsely populated, in the sense that most languages only have annotations for some features, and skewed, in that few features have wide coverage. As typological features often correlate with one another, it is possible to predict them and thus automatically populate typological KBs, which is also the focus of this shared task. Overall, the task attracted 8 submissions from 5 teams, out of which the most successful methods make use of such feature correlations. However, our error analysis reveals that even the strongest submitted systems struggle with predicting feature values for languages where few features are known.
U2 - 10.18653/v1/2020.sigtyp-1.1
DO - 10.18653/v1/2020.sigtyp-1.1
M3 - Article in proceedings
SP - 1
EP - 11
BT - Proceedings of the Second Workshop on Computational Research in Linguistic Typology
PB - Association for Computational Linguistics
T2 - The 2020 Conference on Empirical Methods in Natural Language Processing
Y2 - 16 November 2020 through 20 November 2020
ER -
ID: 254997462