Model of automated system providing information search on a specific topic on the Internet

Abstract

Currently, there is a need to address a number of issues related to the functioning of information retrieval systems. Main among them is the development of algorithms and methods to efficient search and further process information on a particular topic and, as a result, to create an automated information retrieval system (AIS) on a particular subject on the Internet. AIS should maintain an archive of queries, include a thesaurus, spell checking and parsing tools for language queries. To increase the completeness of the search, a thesaurus can be used, where the words related to the query are added together with the similar words from a thesaurus. The future AIS should take into account the work with various Internet services used by special services analysts, for example, search engines, thematic resource catalogs, news sites, RSS messages and news agencies that broadcast news online. For this purpose, the first step in the design of the ASPI is to create a database of search resources available on the Internet, taking into account their specific features in providing information on a particular subject. To improve search, it is planned to use thematic classification of resources - vectors of the vocabulary space (indexing terms) of a system. The task will be to select best possible features and formulate rules on the basis of which a decision will be made regarding a resource of a certain subject. Further work of ASPI is related to processing material found . To solve this task, a necessary step in creating a future system is to organize automatic abstracting of found information. To visualize the found information for purpose of its further analysis, it is advisable to use technology of building semantic networks. Comparison between semantic networks of different texts allows establishing the degree at which they are semantically similar, which can be used for automatic classification according to specified headings, searching for documents based on similarity of a given text, and dividing information array into classes containing similar content. In accordance with above approaches, this article proposes a schematic structure of model AIS on a specific topic on the Internet. The use provided by proposed model of an automated system for searching thematic information on the Internet during information and analytical activities in special units will provide an analytical operator with means for quick and efficient search of heterogeneous information on monitored objects, quick identification of implicit links between monitored objects and related facts and events; recording and visualizing results of analytical research by generating digests of articles, facts, formalized dossiers, semantic networks, and other analytical reports. Key words: model, Internet, automated information retrieval System, information request.