Botnet Attacks Detection using Image Processing Approach: Sequential Neural Network and Feature Selection
Taufiq Odhi Dwi Putra, Tohari Ahmad · 2023
Botnets are networks of compromised computers that are controlled by an attacker for malicious purposes such as spamming attacks, DDoS, or data theft. Botnet attacks detection is a challenging task due to their distributed and covert nature. Botnet attacks are a major security concern for organizations and individuals alike. In this paper we proposed a botnet attacks detection using image processing approach. We propose an AI model architecture using Sequential Neural Network and feature selection. There are two methods for converting raw data into image data. First method using predefined categorize for choosing features. The second method uses Pearson correlation value between features in dataset and the label. Predefined categorized features have successfully increased the performance of proposed AI model for botnet attacks detection as well as Pearson correlation value.