Abstract:
The paper proposes a method for automatic construction of a
scientific terminology thesaurus based on algorithms for extraction of
multi-word terms from special encyclopedias and scientific publications.
The results of the algorithms for thesaurus creation and replenishment
are presented on the example of mathematical text processing.
We propose the algorithm for comparative semantic analysis of
scientific publications and the ways of quantitative estimation of their
semantic similarity.
Keywords:aspect-oriented analysis, scientific vocabulary, semantic
graph, classification of scientific text, automatic processing of unstructured texts