Finding Cognate Groups Using Phylogenies
David Hall, Dan Klein · 2010
A central problem in historical linguistics is the identification of historically related cognate words. We present a generative phylogenetic model for automatically inducing cognate group structure from unaligned word lists. Our model represents the process of transformation and transmission from ancestor word to daughter word, as well as the alignment between the words lists of the observed languages. We also present a novel method for simplifying complex weighted automata created during inference to counteract the otherwise exponential growth of message sizes. On the task of identifying cognates in a dataset of Romance words, our model significantly outperforms a baseline approach, increasing accuracy by as much as 80%. Finally, we demonstrate that our automatically induced groups can be used to successfully reconstruct ancestral words. 1
Cited by 37 works.
Authors
References
- The Hungarian method for the assignment problem
- Expectation Propagation for approximate Bayesian inference
- Inducing multilingual text analysis tools via robust projection across aligned corpora
- Weighted Automata in Text and Speech Processing
- Philological Probability Problems
- Parameter estimation for probabilistic finite-state transducers
- Indo‐European and Computational Cladistics
- Multipath translation lexicon induction via bridge languages
- First- and second-order expectation semirings with applications to minimum-risk training on translation forests
- Cognates can improve statistical translation models