Исследование частотного распределения анафоры в разноуровневых адаптированных русскоязычных текстах как приложение к задаче автоматического разрешения референции

Abstract

The article considers the problems of automatic anaphora resolution in Russian texts and provides a brief description of reference classifications. The analysis of frequency distribution of model anaphora is performed on the training materials suggested to the international students within the learning process by the department of Russian as Foreign, TPU. The course material given in the study covers the following knowledge levels: elementary level (A1), basic level (A2) and first certification level (B2). The purpose of the given article is to provide statistical assessment of these or those types of anaphora depending on the level of the material considered, as well as to determine what types are more common and to indicate the future direction in the study of automatic reference resolution problems of Russian texts.