Predicting Typological Features in WALS using Language Embeddings and Conditional Probabilities: ÚFAL Submission to the SIGTYP 2020 Shared Task

Martin Vastl, Daniel Zeman, Rudolf Rosa


Abstract
We present our submission to the SIGTYP 2020 Shared Task on the prediction of typological features. We submit a constrained system, predicting typological features only based on the WALS database. We investigate two approaches. The simpler of the two is a system based on estimating correlation of feature values within languages by computing conditional probabilities and mutual information. The second approach is to train a neural predictor operating on precomputed language embeddings based on WALS features. Our submitted system combines the two approaches based on their self-estimated confidence scores. We reach the accuracy of 70.7% on the test data and rank first in the shared task.
Anthology ID:
2020.sigtyp-1.4
Volume:
Proceedings of the Second Workshop on Computational Research in Linguistic Typology
Month:
November
Year:
2020
Address:
Online
Venues:
EMNLP | SIGTYP
SIG:
SIGTYP
Publisher:
Association for Computational Linguistics
Note:
Pages:
29–35
Language:
URL:
https://aclanthology.org/2020.sigtyp-1.4
DOI:
10.18653/v1/2020.sigtyp-1.4
Bibkey:
Cite (ACL):
Martin Vastl, Daniel Zeman, and Rudolf Rosa. 2020. Predicting Typological Features in WALS using Language Embeddings and Conditional Probabilities: ÚFAL Submission to the SIGTYP 2020 Shared Task. In Proceedings of the Second Workshop on Computational Research in Linguistic Typology, pages 29–35, Online. Association for Computational Linguistics.
Cite (Informal):
Predicting Typological Features in WALS using Language Embeddings and Conditional Probabilities: ÚFAL Submission to the SIGTYP 2020 Shared Task (Vastl et al., SIGTYP 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.sigtyp-1.4.pdf
Video:
 https://slideslive.com/38939792
Code
 ufal/ST2020