The Algorithm of Semantic Analysis in Disruptive Information Security Systems

Contribution

In the framework of this study, we propose an experimental study of the algorithm, with the following improvements:

• · the possibility of correction of data indexing depending on its location in the text;

• · the categorization of the studied data sets by structured dictionaries with a specific signature database

The formulation of the scientific hypothesis is as follows: suppose that a given frequency of occurrence of tokens from signature bases prepared by the organization will more qualitatively indicate a certain level of its information security depending on the categorization of signature bases and correction for its location in the text.

The problem, the solution of which is proposed to be obtained in the work, is defined as optimization, namely: the need to maximize the number of fragments found, that informatively for the researcher cross the perimeter of the given signature constraints while minimizing the errors of false positives.

Thus, new parameters are added that determine the quality of the “output” of the processed information, which allows the end user to quickly find, or skip less the areas of actual interest without the additional working time expenditures for “false” alarms (pieces of information that actually do not indicate anything).