Statystyczne właściwości tekstów prawnych i ich wykorzystanie w systemach wyszukiwania informacji prawnej

The article is devoted to issues related to the use of statistical properties of legal texts for searching legal information. Methods are presented that allow to enlarge the set of found documents by including those semantically close to the initially found ones. Providing the ability to search for such collections of documents allows to better satisfy the needs of the system users. The article presents the theoretical foundations of enlargement operations based on performing specific treatments in the so-called semantic space of terms and semantic space of documents. The article deals particularly with the methods which allow to determine the set of similar documents by using statistical properties of documents. The research carried out under the ERASMUS program conducted by K. van Noortwijk and R. V. de Mulder is presented in detail. A critical analysis of the measures used in this research was carried out, as well as an analysis of the reasons why the proposed methods didn’t lead to fully satisfactory results.