A.V. Pavlenko – Employee, Cherepovets Higher Military Engineering School of Radio Electronics
The huge volumes of text information transmitted in media and electronic text storages call into existence of it automated processing to gain different objectives. The quality of text processing systems depend on informative text representation. Methods based on functionally-role interpretation allow creating more informative text representation than the representation consisting only of keywords, but are very resource-intensive. The main idea of the proposed method is to avoid resource-intensive procedures, such as semantic analysis. The FRI is constructed on the morphological data and POS-tagging results supplemented by heuristic rules set. The proposed formalization method dependency between text message volume and its formalization time is linear instead the exponential dependency in other methods.
- Stolyarov M.G., Novikov A.Yu. Method of text information ranging in case of full-textual search using relations between terms // Knowledge-intensive technologies. 2012. V. 13. № 8. P. 87−90.
- Zelenkov Yu.G., Segalovich I.V., Titov V.A. Probability POS-tagging model based on normalizing substitutions and nearest words positions // Materials of international «Dialog» conference. 2005. URL = http://www.dialog-21.ru/media/2444/zelenkov_segalovich.pdf.
- Mann W.C., Thompson S.A. Rhetorical structure theory: Toward a functional theory of text organization // Text. 1988. № 8(3). P. 243−281.
- Golikov I.Yu. Features of elementary and subsequent processing of big unstructured textual data // Materials of IV science-technical conference «RTI – Antiaircraft defense systems-2016». 2016. P. 520−527.
- Mitelkov D.V., Novikov A.Yu. Metod opredeleniya informacionnoj cennosti tekstovyh soobschenij // Science intensive technologies. 2016. V. 17. № 12. Р. 67−70.