• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site

Book

Труды МФТИ

Т. 9. Вып. 2 (34). М.: 2017.

 

In this paper, the main purpose is to consider applications of morphological analysis in
text classifiation. Morphological analysis helps us to learn grammatical features of words,
grammatical semantic and the interaction between the elements of text. We propose the
neurosemantic network based on morphological analysis for learning vector representations
of the text’s grammatical structures and the recursive autoencoder that consists of two
parts - the fist part combines two vectors of words, the second one combines two vectors of
morphology.
 








Труды МФТИ