Deep learning techniques to detect Botnet

Abstract

During the economic and social development of the country, digital transformation is an indispensable requirement. Information and network security are two major impediments to successful digital transformation. Cyberattacks are happening every day, every hour and in fact, Vietnam is among the countries most attacked by cyberattacks in the region but has the lowest cybersecurity index. In the past time, the world has witnessed an unprecedented explosion of deep learning. Besides the development of information technology, security and safety threats are also increasing, one of which is the Botnet network. Because Botnet networks are becoming more complex and difficult to detect, traditional techniques are no longer effective, and one of the most pressing issues today is to find an effective solution for detecting botnets. Based on the advantages of deep learning such as scalability, performance and execution time, interpretability, etc., the authors have installed and evaluated 3 methods of Botnet detection. The results obtained are outstanding. Therefore, in this paper, the authors have used the deep learning technique and proposed to build a 3-layer neural network model for detecting and warning against attacks using Botnets. Through comparison and evaluation, the results obtained by the proposed neural network model were better than other methods such as SVM, RNN, and LSTM.