In her work the author considers methods for automatic extraction of abstracts from documents. It begins with formulation of a problem, introduction of basic concepts and overview of the literature on the current topic. Then the choice of the method based on the clustering of the corpus of documents is justified and the methods of clustering are considered. Then all the elements are assembled into a single algorithm and tested on real data.  One can note some incompleteness of the research description. For example, it was worth noting the features of constructing abstracts in the case of a very large number of clusters or only one cluster. It would also be interesting to look at the result of abstracting a rather large article.  This incompleteness, however, is rather an opportunity for the further development of work than a significant drawback. The result of the work was a ready-made solution to the problem of automatic abstracting. The solution was obtained not by writing code for a single well-known algorithm, but as a result of theoretical and practical research. The work turned out to be interesting and deserves an excellent grade.