Concept list Petroni 2010 100

The present list was created in the development of methods of automated reconstruction of language relationships, tested on the Indo-European languages. It was initially based on Dyen's 200-item dataset, identical to Swadesh's list (1952), but, for the purpose of the study, was reduced to 100 concepts with the highest stability measures. As pointed out by the authors, the list does not make use of any historical information such as cognacy. Instead, it is designed to compare words with the same meaning for computing lexical distances between lexemes and establishing relationships among languages without the need for historical input.

Id English Concept set