Linguistic Processor Semantix for Knowledge Extraction from Natural Texts in Russian and English.
Igor P. Kuznetsov, Elena B. Kozerenko · International Conference on Artificial Intelligence · 2008
The linguistic processor Semantix is intended for the areas where the automatic formalization of the flows of texts in natural language is required: resume, mass media issues, information and advertising materials, mail communications, summaries of incidents, information in the criminal cases, archive materials and other texts. The objects interesting for a user are extracted from documents with their features and relations. As a result on the basis of each document a special form of the semantic network is built, which reflects its semantic structure. Such networks are mapped onto the XMLfiles, which serve for organizing the bases of knowledge, corresponding to semantic search, for the solution of logical analytical problems, and also for the automatic filling of relational databases