Botnet Detection Based on Generative Adversarial Network and Efficient Lifelong Learning Algorithm

Abstract

Aiming at the problem that traditional machine learning cannot apply learned knowledge to new tasks and has poor classification effect when dealing with unbalanced botnet detection data, This paper proposed a botnet detection method based on generative adversarial network(GAN) and Efficient Lifelong Learning Algorithm(ELLA). Firstly, the time window is used to extract the feature of the CTU-13 data set. Then, the training set and the test set are divided, and the GAN is used to expand the data of a small number of samples in the training set. The ELLA model is trained using the expanded training set. Finally, experiments are carried out on the test set. The experimental results show that compared with B-ELLA method, the average accuracy is increased by 8.15 %, and the average recall rate is increased by 14.5 %. Compared with the traditional machine learning method, GAN-ELLA can also achieve better detection results.