In the paper, problems of legal information digitalization are investigated. Conditions for extraction information from legal texts related to the common ones processing (non-legal terms) are outlined. Sample results of similarity analysis are presented. Further research aimed at semantic analysis of legal texts are outlined.