The Algorithm of Semantic Analysis in Disruptive Information Security Systems

The Algorithm of Semantic Analysis in Disruptive Information Security Systems

Polina Kisarina, Andrey Mishin
Copyright: © 2021 |Pages: 13
DOI: 10.4018/978-1-7998-0361-4.ch007
OnDemand:
(Individual Chapters)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

This chapter represents some of the main drawbacks of DLP systems implemented by businesses in international practice. The main structural shortcomings of these systems have been analyzed, and the factors correlating with them were revealed. An experimental setup has been formed to assess the impact of changes in these factors on the Type 1 and 2 errors in the operation of the systems. The authors also provide the results of the research with the use of algorithms, including the influence of the identified factors in the business systems of different directions to improve the economic security of the company.
Chapter Preview
Top

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).

Complete Chapter List

Search this Book:
Reset