Final qualifying work of Diurdevа Polina has important topic, because currently it becomes important to use automatic methods of identifying the author of  documents. Some methods require additional analysis due to the diversity of potential training data sets and purposes of use of the methods. In the work Diurdeva P. analyzes an applicability of the method which is based on the distribution of frequencies of letter combinations.
	Final qualifying work consists of six chapters and an introduction, conclusion and list of references. References to the used literature are provided correctly.
	In the introduction  the relevance of the research is justified,  the problem statement and aims of the work are presented.
	In the first Chapter the tasks necessary to achieve the aims of the work are stated.
	In the second Chapter existing similar systems are described in terms of their approaches to solving the problem of identification of the author, conditions on the input data, the recognition accuracy. 
	In the third Chapter the investigated method of author identification is described. Description of the main mathematical tools needed to build a system is also provided in this chapter. 
	In the fourth Chapter the dataset which was used for testing the implemented by Diurdeva P. system for author identification is described.
	In the fifth Chapter the results of the classification of texts by author's affiliation is described. Moreover, analysis of the obtained results is presented, the dependencies of accuracy on various conditions on the input data are investigated.
	In the sixth Chapter the results of applying the studied method for solving the problem of text clustering by author are described.  Also, algorithms and metrics using which testing and analysis were made are described. Corresponding conclusions are made.
	In the final chapter of qualifying work the conclusions about the applicability of the investigated approach for solving the problem of author identification are presented.
	The research of Diurdeva Polina showed that the investigated method can be used to solve important task of authorship attribution of a document that is important in linguistic, historical and forensic studies, and can also be used in systems for detection of plagiarism.
	Final qualifying work of Diurdeva P. may be admitted for defence. Recommended rating: "excellent".