Abstract
Within the framework of the CoToHiLi project (Computational Tools for Historical Linguistics), set up with the aim of providing computational tools for the research of linguistic change, we created a database containing the cognate chains from the main Romance languages. Starting from this database, we conducted experiments that test the machine’s ability to automatically detect cognate sets in any pair of Romance languages. We also explored the possibility of distinguishing semi-automatically between inherited and borrowed words in the Romance languages. The article describes the challenges of each experiment, the application of digital methods, as well as the importance of integrating linguistic data. We also propose an interpretation of the results.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have