Registration Number
0126U005083,
R & D request
Title
Development of a Methodology for Improving the Security of Web Applications through Deep Learning
Head
Iryna V. Zamrii, Доктор технічних наук
Registration Date
19-10-2026
Organization
Zamriі Iryna Viktorivna
popup.description1
The aim of this work is to develop a methodology for enhancing web application security through the application of deep learning, thereby ensuring the efficient operation of enterprises and fostering the country's economic development. This entails creating adaptive systems that leverage the capabilities of neural networks to automatically detect and counter modern cyber threats, such as web spam and DDoS attacks. To achieve this goal, the following tasks are addressed: conducting a comprehensive analysis and comparative study of the effectiveness of modern technologies versus traditional web application security methods; researching and developing a model based on graph neural networks for the high-precision detection and classification of web spam; detecting DDoS attacks to enable system adaptation to traffic fluctuations without the need for constant rule updates; implementing a three-level decision-making system that balances automation and control; reducing system load; minimizing false positives; and ensuring the self-learning and adaptation of models based on new data and feedback, allowing the system to respond promptly to emerging threats and maintain a high level of web application security.
popup.nrat_date
2026-10-19
