Jambu: A historical linguistic database for South Asian languages
This provides a structured and accessible resource for historical linguists and Indologists, though it is incremental as it builds on existing sources.
The researchers tackled the problem of fragmented historical linguistic data for South Asian languages by introducing Jambu, a unified cognate database that consolidates 287k lemmata from 602 lects into 23k cognate sets, and they trained neural models for reflex prediction on its Indo-Aryan subset.
We introduce Jambu, a cognate database of South Asian languages which unifies dozens of previous sources in a structured and accessible format. The database includes 287k lemmata from 602 lects, grouped together in 23k sets of cognates. We outline the data wrangling necessary to compile the dataset and train neural models for reflex prediction on the Indo-Aryan subset of the data. We hope that Jambu is an invaluable resource for all historical linguists and Indologists, and look towards further improvement and expansion of the database.