Saved in:
Bibliographic Details
Main Authors: Chang, Kalvin, Robinson, Nathaniel R., Cai, Anna, Chen, Ting, Zhang, Annie, Mortensen, David R.
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.01582
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914664449835008
author Chang, Kalvin
Robinson, Nathaniel R.
Cai, Anna
Chen, Ting
Zhang, Annie
Mortensen, David R.
author_facet Chang, Kalvin
Robinson, Nathaniel R.
Cai, Anna
Chen, Ting
Zhang, Annie
Mortensen, David R.
contents We describe a set of new methods to partially automate linguistic phylogenetic inference given (1) cognate sets with their respective protoforms and sound laws, (2) a mapping from phones to their articulatory features and (3) a typological database of sound changes. We train a neural network on these sound change data to weight articulatory distances between phones and predict intermediate sound change steps between historical protoforms and their modern descendants, replacing a linguistic expert in part of a parsimony-based phylogenetic inference algorithm. In our best experiments on Tukanoan languages, this method produces trees with a Generalized Quartet Distance of 0.12 from a tree that used expert annotations, a significant improvement over other semi-automated baselines. We discuss potential benefits and drawbacks to our neural approach and parsimony-based tree prediction. We also experiment with a minimal generalization learner for automatic sound law induction, finding it comparably effective to sound laws from expert annotation. Our code is publicly available at https://github.com/cmu-llab/aiscp.
format Preprint
id arxiv_https___arxiv_org_abs_2402_01582
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Automating Sound Change Prediction for Phylogenetic Inference: A Tukanoan Case Study
Chang, Kalvin
Robinson, Nathaniel R.
Cai, Anna
Chen, Ting
Zhang, Annie
Mortensen, David R.
Computation and Language
We describe a set of new methods to partially automate linguistic phylogenetic inference given (1) cognate sets with their respective protoforms and sound laws, (2) a mapping from phones to their articulatory features and (3) a typological database of sound changes. We train a neural network on these sound change data to weight articulatory distances between phones and predict intermediate sound change steps between historical protoforms and their modern descendants, replacing a linguistic expert in part of a parsimony-based phylogenetic inference algorithm. In our best experiments on Tukanoan languages, this method produces trees with a Generalized Quartet Distance of 0.12 from a tree that used expert annotations, a significant improvement over other semi-automated baselines. We discuss potential benefits and drawbacks to our neural approach and parsimony-based tree prediction. We also experiment with a minimal generalization learner for automatic sound law induction, finding it comparably effective to sound laws from expert annotation. Our code is publicly available at https://github.com/cmu-llab/aiscp.
title Automating Sound Change Prediction for Phylogenetic Inference: A Tukanoan Case Study
topic Computation and Language
url https://arxiv.org/abs/2402.01582