Searching Legal Information Using a Natural Language
Searching Legal Information Using a Natural Language

Summary/Abstract: The article deals with a problem of constructing computer retrieval systems based on the use of natural language. Those kind of systems differ from the classical ones and relay on the idea that the user should have the possibility to introduce to the system search instructions in the same manner in which she or he asks a question. Searches in such systems don’t use any retrieval languages. They don’t use Boolean operators and every single word from the instruction is used as a separate search criterion. To the retrieved documents weights are assigned on the basis of statistical analysis of the text of the documents. This allows the ranking of documents and presenting the results of the search in an ordered manner. The first attempts of using such a system took place in 1960s, but in the field of legal information from the beginning of 1990s. The article presents the methods of searching used in WESTLAW IS NATURAL (WIN) created by West Publishing Company, FREESTYLE SEARCHING developed by Lexis Nexis and WESTLAW NEXT based on the idea Google for lawyers. The critical analysis of functioning and estimation of retrieval results shows that those kind of systems cannot replace traditional ones based on Boolean searching but can sometimes play a supplementary role.

• Details
• Contents