Computing Russian Morphological distribution patterns using RusAC Online Server

Abstract

The article presents findings of distribution patterns of Russian grammatical categories computed with the help of MyStem.3 tagger and a proprietary Russian language processor, ETAP-3. The corpus of over 1.1 mln tokens compiled for the study comprises two types of academic texts used in Russian schools: Science textbooks and Humanities textbooks. We computed descriptive metrics of each textbooks with the help of the text analyzer RusAC (http://tykau.pythonanywhere.com/) and pursued the contrastive analysis of Science and Humanities classroom textbook features. Significant differences of two types of the texts were found in distribution patterns of noun cases and verbs tenses, while morphological patterns of nouns, adjectives, verbs, adverbs demonstrate similarities. The specifics of grammatical patterns defined for classroom textbooks can be used in further studies on distribution of morphological patterns and text complexity of Russian academic texts