Method for detecting botnets through deep training developed

The next joint scientific seminar "Development of botnet detection method with Deep Learning based on C & C communication images" of departments No.1  and 2 was held at the Institute of Information Technology of ANAS.

The Ph.D. student of the institute, Gulnara Garayeva, mentioned principles of organizing a command-management system in the botnets with the malware-infected computers. She pointed out that some of the characteristic features of the data exchange between the command-management centers and boots permit the detection of botnets, and in recent years research on botnets has been applied to artificial intelligence and deeper learning methods.

The speaker gave a broad overview of various neuronal networks, widely used in various fields, including recurrent neuronal networks, autoencoder and convolutional neural networks.

At the seminar, the deep neural network architecture proposed for this issue was justified and its functions were explained in detail. The key features of databases used in experiments, and information on botnets used for recognition of botnets were given and the results of experiments conducted with various hyperparameters in the proposed architecture were highlighted.

In the end, views on presentation were exchanged and questions were answered.

© All rights reserved. Citing to www.ict.az is necessary upon using new