Leveraging AI and Network Forensics for Smart Botnet Detection

Abstract

Nowadays, ensuring security for computers, networking devices and handheld devices has become a major concern, as numerous security threats are evolving rapidly and creating serious security issues. Botnet belongs to one such threat that is creating security concerns. Botnet is a collection of hijacked devices managed by a botmaster that are used to launch massive cyber threats. As cybercriminals aim to target users’ computer networks, they have grown stealthier and more adaptive, often slipping past conventional security systems undetected. Devices that get connected to a botnet consist of systems, handheld devices like mobiles, and Internet of Things devices. The proposed approach introduces a hybrid behavioral–temporal framework for detecting bots that are compromised through live traffic monitoring. The traditional approaches are more dependent on publicly available datasets, whereas the proposed framework, instead of relying on such methods, utilizes live or real-time traffic, which enables to find the bots in real operational conditions. This framework utilizes multidimensional behavioral analysis to find the various activities associated with the botnet. By integrating real-time anomaly detection, it helps in finding both rapid and stealthy low-rate infectious activities of the botnet. Nevertheless, the framework eliminates the need for deep packet inspection, which is more suitable for scalable and even encrypted traffic networks. When this framework is compared to conventional methods, it achieves high precision, recall, and F1-score values while maintaining low false-positive rates, making it a highly effective tool for real-time threat detection.

Received: 9 April 2026 | Revised: 10 July 2026 | Accepted: 30 August 2026

Conflicts of Interest

The authors declare that they have no conflicts of interest to this work. 

Data Availability Statement

The data that support the findings of this study are openly available in Figshare at https://figshare.com/s/c5c40c16d0f343a2bf31. 

Author Contribution Statement

Shivanna Doddegowda: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization. Pooja Mysore Raghupathi: Conceptualization, Methodology, Investigation, Resources, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration.