Аннотация
The paper considers an algorithm that allows us to automatically construct a list of characteristic suffixes by the dictionary of stems with prescribed morphological classes that are at our disposal. A set of heuristic rules of rejecting noninformative stems and cutting off the remaining informative stems is proposed starting from the comparison of similarity of neighboring stems in the dictionary with the inverse alphabet order. The results of effectivity of work of the proposed algorithm are presented for the Russian, Ukrainian, and French languages. Ключевые слова
characteristic suffixes, morphological classes, inflective language |