COMPARATIVE ANALYSIS OF NON-FINITE FORMS AND AUTOMATIC TEXT PROCESSING

Abstract

The article analyses the non-finite model learned from texts of the Internet resource http://www.tabasaran.webonary.org/is a national portal providing different fragment inflections. The distribution criterion description of text units with paradigmatic and syntagmatic set gives the ability to distinguish between parts of speech, private values combined opposition verb paronymy. When selecting fragments of national texts with the opposition used patronymics solid method verb fetch tokens corresponding to the tasks of the Corps study of the national language. The specificity of annotating forms of the verb, systematizing the markup language Corpus indicates insufficiently known aspect of literary translation. A large part of the categorical formants of literary translation is submitted to the national texts. The first step of processing national Corpus of texts carried out presentation of entries-infinitive, gerundive, participle, Masdar.