This paper presents the results of a research study that explores the methods of analysis of unstructured text information. A new method that is applied to determine the similarity of scientific and technological documents based on the thematic significance feature is proposed. The results that were obtained using several modifications of the existing methods are compared experimentally.
Suvorov R. E., Sochenkov I. V. Establishing the similarity of scientific and technical documents based on thematic significance // Scientific and Technical Information Processing. – 2015. – Т. 42. – №. 5. – С. 321-327.