Reinforced machine learning methods for testing quality of cyber threat prediction results

Authors

  • A.B. Kachynskyi National Technical University of Ukraine «Igor Sikorsky Kyiv Polytechnic Institute», Institute of Physics and Technology, Ukraine
  • N.A. Tsebrinska National Technical University of Ukraine «Igor Sikorsky Kyiv Polytechnic Institute», Institute of Physics and Technology, Ukraine

DOI:

https://doi.org/10.20535/tacs.2664-29132020.1.209432

Abstract

The article considered on machine learning methods with reinforcement to make decisions about evaluating the quality of a mathematical prediction model. Given the problems of cybersecurity specificity A/B testing algorithms, analysis of variance (ANOVA), as well as multi-armed bandit are presented. Features of their practical implementation are taken into account: data type and distribution function, sample size, knowledge about the dispersion of the general population, dependence, and independence of observations. The cybersecurity problems solved with the help of these algorithms are discussed and the methods of their solution are suggested.

Downloads

Published

2020-08-06

Issue

Vol. 2 No. 1 (2020)

Section

Mathematical methods, models and technologies for secure cyberspace functioning research

License

Authors who publish with this journal agree to the following terms:

  1. Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.

  2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.

  3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).