An approach to a cyberattacks management system

Resumen

Cybersecurity is a very necessary area due to the new types of cyberthreats that areconstantly appearing. However, there are not enough tools and techniques to keepsystems safe from cyber-attacks. Therefore, this work aims to develop a system that,based on network trac, not only detects ongoing cyber-attacks in real time but isalso able to predict them suciently in advance so that action can be taken in time toneutralise them. We hypothesize that such a system would make it very dicult forcybercriminals to achieve their goals and would reduce the risk of cyberattacks. Thisgoal could be achieved by (1) selecting a suciently good dataset, which contains awide variety of the most current and popular types of attacks, and contains enoughdata to be able to build a good system; and (2) using dierent articial intelligencetechniques. The use of these techniques requires sucient data to train a successfulmodel for the system to be able to predict well enough. Therefore, it will also benecessary the design of a network trac simulator to inject both benign and malignnetwork packets.