Application of text mining technologies in Russian language for solving the problems of primary financial monitoring

The paper deals with the issues of text mining based on a vector classification model. The term frequency and inverse document frequency function (TF IDF) is used as a measure to evaluate the selection of terms. The text preprocessing stage performs tokenization and normalization of the text. The algorithms of stemming and lemmatization of the Russian-language text are used during normalization. A set of programs has been created that makes it possible to analyze the selected subject area, create a thesaurus for thematic search for violations of the public procurement Federal Law of the Russian Federation, as well as to implement functionality for automated search and analysis of documents of arbitration courts within the framework of this law.