INDEXING FULL-TEXT DOCUMENTS FOR INTELLIGENT KEYWORD INFORMATION SEARCH.

The issue of indexing full-text documents automatically for solving the problem of intellectualizing data searches is considered in the paper. The main objective of the study lies in developing a full-text keyword search model, taking into account morphological features of Russian, as well as in developing algorithms of indexing and a fulltext search. For a practical implementation of the system in the form of a web application, the PHP programming language was chosen, as a relational full-text index database, i.e. DBMS MySQL. For a morphological analysis a "demon'' normalizer, serving as a tcp-server and including the Dialing morph-analyzer, was developed. The given system retains a number of features: it can be used by several users simultaneously, operate great indices, maintain the optimum ratio of selectivity and sensitivity at searching. The research results can be used by analytical linguists, specialists in the field of automation of library activities, as well as other specialists and experts in creating automated library information systems, automatic abstracting systems, etc. Thus, using the abovedescribed software and applications has allowed developing an effective system of indexing full-text documents and full-text keyword searching [ABSTRACT FROM AUTHOR]