Predicting Typological Features in WALS using Language Embeddings and Conditional Probabilities: \'UFAL Submission to the SIGTYP 2020 Shared Task
Computation and Language
2021-10-26 v1 Artificial Intelligence
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.
Cite
@article{arxiv.2010.03920,
title = {Predicting Typological Features in WALS using Language Embeddings and Conditional Probabilities: \'UFAL Submission to the SIGTYP 2020 Shared Task},
author = {Martin Vastl and Daniel Zeman and Rudolf Rosa},
journal= {arXiv preprint arXiv:2010.03920},
year = {2021}
}