2018/04/15 by Taraka Rama, Johann‐Mattis List, Rama, Taraka +5 · 1 citation
Computer Science · Social Sciences · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Language and cultural evolution #Linguistic Variation and Morphology #Natural Language Processing Techniques
paper · pdf · doi:10.48550/arxiv.1804.05416
openalex publication_date 2018/04/15 · openalex created_date 2022/10/05 · openalex updated_date 2026/07/28
We evaluate the performance of state-of-the-art algorithms for automatic\ncognate detection by comparing how useful automatically inferred cognates are\nfor the task of phylogenetic inference compared to classical manually annotated\ncognate sets. Our findings suggest that phylogenies inferred from automated\ncognate sets come close to phylogenies inferred from expert-annotated ones,\nalthough on average, the latter are still superior. We conclude that future\nwork on phylogenetic reconstruction can profit much from automatic cognate\ndetection. Especially where scholars are merely interested in exploring the\nbigger picture of a language family's phylogeny, algorithms for automatic\ncognate detection are a useful complement for current research on language\nphylogenies.\n