• A
  • A
  • A
  • АБB
  • АБB
  • АБB
  • А
  • А
  • А
  • А
  • А
Обычная версия сайта

Статья

QUANTITATIVE EVALUATION OF SYNTAX SIMILARITY

Mathematica Montisnigri. 2019. Vol. XLVI. P. 123-132.
Klyshinskiy E., Karpik O. V.

Machine learning systems are facing problem of incomparability of their results in case of different languages; one of the subarea here is quantitative analysis of syntax. In this paper, we introduce a new quantitative method based on statistics of words co-occurrence in syntactically tagged corpora. The method allows quantitatively evaluate difference and similarity among languages, select most influential phenomena. Experimental setup consists materials for more than 50 languages. Our experiments demonstrate that the introduced method correctly cluster languages among language families.