Peculiarity of lexical-grammatical analysis for object extraction from natural language texts.

Igor Kuznetsov (igor-kuz@mtu-net.ru), Nikolay Somin(somin@post.ru)

Russian Academy of Science, Institute of Informatics Problems

Paper analyses the experience of using the linguistic processor which extracts knowledge (information objects and their links) from natural language texts. Significant part of the processor is the procedure of  lexical-grammatical analysis which many times have been modified in process of tuning on various subject fields. Now the procedure has many peculiarities which are considered in the paper.