Yu. V. Leonova, A. M. Fedotov, O. A. Fedotova, “On the Approach to the Classification of Thesis Abstracts on Themes”, Novosibirsk State University Journal of Information Technologies, 2017, Volume 15, Issue 1,Pages <nobr>47

On the Approach to the Classification of Thesis Abstracts on Themes

Yu. V. Leonova^a, A. M. Fedotov^ab, O. A. Fedotova^bc

^a Institute of Computational Technologies, Siberian Branch of the Russian Academy of Sciences, 6 Academician Lavrentiev Ave., Novosibirsk, 630090, Russian Federation
^b Novosibirsk State University, 2 Pirogov Str., Novosibirsk, 630090, Russian Federation
^c State Public Scientific Technological Library of the Siberian Branch of the Russian Academy of Sciences, 15 Voskhod Str., Novosibirsk, 630200, Russian Federation

Abstract: The method of thematic classification of thesis abstracts is considered in the work. For this purpose, a specially constructed measure of the proximity of documents is used, taking into account the specifics of the subject area. As scales for the definition of a measure, it is suggested to take the characteristics of the structural attributes of the description of the author's abstracts (scientific novelty, provisions to be defended, etc.). The values of the weight coefficients in the formula for computing the proximity measure are determined by the assumed a posteriori reliability of the data of the corresponding scale.

Keywords: classification of thesis abstracts, weight coefficients, measure of proximity, subject area, model of facet classification, classification algorithm.

UDC: 004.9