Decoding Digital Intruders: Machine Learning Strategies for Botnet Identification

Abstract

The word "Robot-Network" is where the term "Botnet" originates. Botnets can harm our devices by creating damage or posing hazards to the computer in the form of viruses, malware, spyware, and other malicious software. Malware is becoming more and more prevalent every day. There is a big concern from malicious apps that transform mobile devices to bots that could be a part of a larger botnet. It is important to detect mobile botnets to keep our devices safe from hackers. The most popular mobile operating system, Android, is rapidly being targeted by malware. Machine learning techniques are implemented to identify botnets in a variety of ways. We describe a Machine Learning based strategy to detect mobile botnets in this study. Low FP (false positive) rates and real identification rates are few goals of the suggested method. Additionally, utilizing the same dataset and multiple machine learning methods, the objective is to enhance identification accuracy. Using various ways to get the best output and accuracy, as well as classifying botnets into families, will help to keep our devices safe from botnets.