FORMATION METHOD OF TEXT FRAGMENTS FOR SEARCH SYSTEMS BASED ON THE DISTRIBUTION OF TERMS BY DOCUMENT

Abstract

Increasing the amount of information needed for people activities in various fields, growing very fast. Finding relevant information for the user is the primary goal of any information system. On the basis scientific papers of general statements text characterized only by the number of keywords in the text. It was impossible to select a fragment of text with a term that is often found. Despite of the fact that a lot of research has been done on this topic, they do not completely solve the problem of formation of text frag- ments. Aim of the paper is the develop the method of formation of text fragments based on the distribution of terms by document. Developed the method of forming text fragments according to the terms of the distribu- tion of the document. Described three approaches installation the ranges and formation of text fragments. Developed the algorithm of fragments formation. Described the principle of selection range for each term. Developed the algorithm of fragments formation for a specific term. Also described the principle of compar- ing several fragments together. Formation method of text fragments for each term can reduce the time for user query. Based on the study can be argued that the finding fragments are an effective way to provide in- formation in comparison with the result as the entire text.