A novel approach to the potentially hazardous text identification under theme uncertainty based on intelligent data analysis

Abstract

The problem of potentially hazardous text identification is an important one in the intelligent data analysis area. As usual, this problem is solved by methods and techniques, which are of a low efficiency in conditions of theme uncertainty. Within this paper, a novel approach to the potentially hazardous text identification under theme uncertainty is presented. The main idea of data processing approach proposed is based on the user and automatically extracted keywords comparison. This paper contains the brief overview of the text identification methods, the description of the approach presented, some statistical experimental results, discussion and conclusion.