Botnets and how to automatic detect them: exploring new ways of dealing with botnet classification

Abstract

Threats such as Botnets have become very popular in the current usage of the Internet, such as attacks like distributed denial of services (DoS) which can cause a significant impact on the use of technology. One way to mitigate such issues can be a focus on using intelligent models that can attempt to identify the existence of Botnets in the network traffic early. Thus, this work aims to evaluate the current state of the art on threats related to Botnets and how intelligent technology has been used in real-world restrictions such as real-time deadlines and increased network traffic. From our findings, we have indications that Botnet detection in real-time still is a more significant challenge because the computation power has not grown at the same rate that Internet traffic. This has pointed out other restrictions that must be considered, like privacy legislation and employing cryptography methods for all communications. In this context, we discuss the following steps to deal with the identified issues.