Cognates are words that have similar spelling and meaning in different languages. They can help a second-language learner on the tasks of vocabulary expansion and reading comprehension. The learner also needs to pay attention to pairs of words that appear similar but are in fact False Friends: they have different meaning in some contexts or in all contexts.
In this talk I will present a method to automatically classify a pair of words as Cognates or False Friends. We focus on French and English, but the method is applicable to other language pairs. We use several measures of orthographic similarity as features for classification. We study the impact of selecting different features, averaging them, and combining them through machine learning techniques.
I will also talk about work in progress on disambiguating Semi-Cognates - pairs of word in French and English that are Cognates in some contexts and False Friends in others.