Performance Analysis of Deep Learning models for DGA domain name detection
Ranjana B Nadagoudar, Dr Sujatha P Terdal
Deep learning, Cyber Security, Domain name generation (DGA), Recurrent Neural Networks (RNN), Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM).
In recent years, the cyber security realm has experienced a notable uptick in reported incidents, particularly concerning malware attacks. These assaults, often orchestrated through botnets networks of compromised devices ranging from computers to IoT gadgets have become go-to tools for cybercriminals. Botnets facilitate a wide array of malicious activities, from DDoS assaults to spam dissemination, data breaches, click fraud, and identity theft. The proliferation of Domain Generation Algorithm (DGA) generated domain names poses a significant challenge in cyber security due to their role in evading traditional detection mechanisms. Traditional intrusion detection systems, reliant on signature-based approaches, find themselves struggling to keep up with the escalating sophistication of botnets. To address this challenge, the proposal suggests harnessing the power of machine learning and deep learning models. The deep learning models, such as Recurrent Neural Networks (RNN), Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM) have shown promise in addressing this challenge. These advanced techniques hold promise in revolutionizing cyber security defense mechanisms, offering a proactive approach to combat the ever-evolving modern cyber threats. However, the demand persists for streamlined models capable of upholding high detection accuracy while conserving computational resources. In this study, we conduct a comprehensive performance evaluation of these models for DGA generated domain name classification and detection. Through extensive experimentation and analysis, we assess the accuracy, precision, recall, and computational efficiency of each model. Our findings provide valuable insights into the effectiveness of RNN, LSTM, and GRU models in achieving high detection accuracy with reduced computational overhead. The proposed model gated recurrent model has outperformed compared to all other deep learning models and achieves high accuracy and detection rate.
Article Details
Unique Paper ID: 163604

Publication Volume & Issue: Volume 10, Issue 11

Page(s): 1840 - 1854
Article Preview & Download

Share This Article

Join our RMS

Conference Alert

NCSEM 2024

National Conference on Sustainable Engineering and Management - 2024

Last Date: 15th March 2024

Call For Paper

Volume 11 Issue 1

Last Date for paper submitting for Latest Issue is 25 June 2024

About Us enables door in research by providing high quality research articles in open access market.

Send us any query related to your research on

Social Media

Google Verified Reviews