Hybrid Double DQN-LSTM Model for Intelligent Botnet Detection in Evolving Cyber Environments

DOI

10.2991/978-94-6239-616-6_111How to use a DOI?

Keywords

Botnet Detection; Double Deep Q-Network; Long Short-Term Memory; Reinforcement Learning; Explainable Artificial Intelligence

Abstract

The rapid expansion of internet-connected devices and IoT infrastructures has significantly increased the scale and sophistication of botnet attacks, making them a major threat to modern digital environments. Traditional signature-based and heuristic detection systems struggle to identify evolving, stealthy, and polymorphic botnet behaviors, often resulting in high false-positive rates and limited resistance to zero-day attacks. To address these challenges, this work presents a Hybrid Double Deep Q-Network–Long Short-Term Memory (Double DQN–LSTM) model designed for intelligent and adaptive botnet detection in dynamic cyber settings. The LSTM component captures temporal and sequential dependencies in network traffic, enabling effective recognition of anomalous behavior patterns that develop over time. The Double DQN component reduces overestimation bias in action-value learning, leading to more stable decisions and improved accuracy in classifying malicious activity. Explainable AI (XAI) methods such as SHAP and LIME are incorporated to provide transparent, human-interpretable insights into model predictions, enhancing trust, interpretability, and operational reliability. By integrating reinforcement learning, deep temporal modeling, and explainability, the approach improves adaptability to evolving attack strategies, reduces false alarms, and supports informed cybersecurity decision making.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - B. Ananth
AU  - P. Sudharsanam
AU  - S. Sriprasath
AU  - S. Adithya
PY  - 2026
DA  - 2026/03/31
TI  - Hybrid Double DQN-LSTM Model for Intelligent Botnet Detection in Evolving Cyber Environments
BT  - Proceedings of the International Conference on Artificial Intelligence and Secure Data Analytics (ICAISDA 2025)
PB  - Atlantis Press
SP  - 1540
EP  - 1551
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-616-6_111
DO  - 10.2991/978-94-6239-616-6_111
ID  - Ananth2026
ER  -