Computational Linguistics: A phonetics driven comparative approach

About Computational Linguistics: A phonetics driven comparative approach

We use edit distance over phonetic descriptions of key words to derive pairwise distances, and then apply advanced data analytics techniques to analyse the relationships between sets of human languages. Our approach is highly computationally efficient, enabling us to analyse very large language sets comprising thousands of members. We use several different sets of words, including basic numbers, colours and key concepts, as well as applying our approach to dialect sheep counting systems.