Abstract:
The semantic-oriented linguistic processor which extracts the knowledge from natural language texts and forms the structures in the Knowledge Base is considered. Significant part of this processor is the block of lexical morphological analysis including the subject vocabularies. The paper discusses the direction of development the vocabularies which are used to analyze the context of words and assigns new semantic signs for them. It gives the possibilities to eliminate various ambiguities in the process of syntactical semantic analysis.
Keywords:knowledge extraction; linguistic processor; lexical morphological analysis; subject vocabulary.