Models for detecting anomalies in the operation of computer security and defense systems using machine learning methods

1. Higher education: 5. Technical sciences

2. Professional field: 5.2 Electrical engineering, electronics and automation

3. Doctoral program: Automated systems for information processing and management.

4. Form of training: part-time (officer, civil servant or citizen outside the system of the Ministry of Defense, SPAM and BA)

5. Applicant: Defense Institute "Professor Tsvetan Lazarov"

6. Primary Unit: Communications and Information Systems and Information Protection Department

7. Accepting structural unit: Development of C4I Systems Directorate

8. Actuality and Disserability of the Suggested Scientific Problem:

Cyber ​​threats are becoming increasingly serious, targeting any type of tactical or strategic object. Attacks can be made from anywhere in the world to anywhere in the globe. Cyber-terrorism is at the highest level and NATO and EU countries need to improve their systems by using 21st century technologies to mitigate cyber threats to military systems, platforms and missions in general. Machine Training (ML) is one of the most advanced technologies that can be used to improve strategic cyber positioning and create the necessary protection that not only addresses today's threats but also threats after 10 and 20 years.

The main goal of the dissertation research is to consolidate the knowledge of machine training and cyber defense, identify gaps between different solutions and needs, and present models based on these military applications.