Author/Authors :
aly, a. a. minia university - faculty of science - computer science department, Egypt , girgis, m. r. minia university - faculty of science - computer science department, Egypt , abdel-latef, b. a. minia university - faculty of science - computer science department, Egypt , el-gamil, b. r. minia university - faculty of science - computer science department, Egypt
Abstract :
There is no doubt that document representation is of great importance to any one dealing with the problem of information retrieval (1R). The question has always been about how we select tokens for the IR system, and on which basis and discipline this selection should be made. This paper tries to answer this question through the introduction of algorithms for token selection based on percentage amount from the actual document.
