Method of thematic clustering of text documents in the base of there infological processing

Proposed method of thematic clustering of text documents based on infological processing of anthology of cluster and new document with purpose to determine correspondence of the terms and calculation of correspondence coefficient. Determination of correspondence of the terms and there promptness requires structural decomposition of text, considering the operations of normalization and identification of the hierarchy of his concepts, comparison of hierarchy of concepts of text with hierarchy of concepts of anthology of cluster and calculation of correspondence coefficient. The decision on the compliance of semantic content of new document and clusters made by evaluating of quantitative values of that coefficient. In turn threshold level of correspondence coefficient determined by experimental paths for different types of text documents.