This article is cited in 4 papers

Natural language processing

Method for detecting relationships between sci-tech documents based on topic importance characteristic

R. E. Suvorov, I. V. Sochenkov

Institute for Systems Analysis of Russian Academy of Sciences

Abstract: The article covers research in the field of Natural Language Processing. The method and the algorithm for searching thematically similar documents are presented. A comparison of various measures of thematic similarity and sets of features is performed.

Keywords: text similarity, vector space model, TF, IDF, topic importance characteristic, measure of thematic similarity, assessment of methods for information retrieval, DCG.

 English version: , 2015, 42:5, 321–327

Bibliographic databases:

•