ავტორი:Manana Khachidze, Magda Tsintsadze, Maia Archuadze, Gela Besiashvili
ინსტიტუტი:Department of Computer Sciences, Iv.Javakhishvili Tbilisi State University, Georgia
In the proposed article the difficulties of Georgian language based semantic retrieval “engine” algorithmdevelopment issues are being considered. The detailed research work plan on Georgian Language web based Corp development/update system for general using purposes is provided. The Corp will be serving as general source for Georgian language based search engines. It is offered to develop modified labeling for non-structured documents using method of Analytical Heuristics in order to form the concept “patterns” knowledge-base. This method of indexing will allow presentation of non-SWD documents in form of “pseudo-SWD” documents and little external engine on SWSS will give an opportunity to develop the semantic retrieval algorithm based on concept patterns categorized by various subject ontologies. We suppose the best online retrieval system in future will be the one oriented on semantic web, thus the natural language Corp based semantic search systems are of uncontested perspective. Index Terms: information retrieval, semantic search, semantic web.