Within the framework of the project KP-06-N57/4 "Research and application of machine learning algorithms in the analysis and development of highly secure software," funded by the Scientific Research Fund, with coordinator for Sofia University, prof. Milen Petrov (Faculty of Mathematics and Informatics), two experimental prototypes were created to address current cybersecurity issues – the detection of phishing sites and the protection of unmanned aerial vehicles from network attacks.
In the first area, an innovative method based on artificial intelligence and machine learning was developed and applied in a new browser plug-in called NotPwned. It helps users recognize and block fake websites created to steal personal data. Additional tools for data collection and processing have also been developed, along with an intelligent system for recognizing login forms and popular brand logos via screen image analysis. A comparison of leading object recognition technologies has also been conducted, with results showing that the developed solution can successfully compete with existing approaches and provide reliable online protection.
The second area of research focuses on the security of drones, whose use in the modern world is constantly growing. Their network's resilience to the most common cyber threats was analyzed, a model of potential attacks was developed, and controlled tests were conducted to identify vulnerabilities. The results obtained are carefully documented and analyzed, and on their basis, specific recommendations are formulated to improve the security of unmanned systems and limit the risk of abuse.
For more information you can visit project web page: https://devops.tu-sofia.bg/ (with mirror on: https://devops2.w3c.fmi.uni-sofia.bg/ ).
