Back

Performance Evaluation of Deep Learning Models for Sequence-Based Intrusion Detection.

  • Published In: International Journal on Electrical Engineering & Informatics, 2025, v. 17, n. 1. P. 63 1 of 3

  • Database: Academic Search Ultimate 2 of 3

  • Authored By: Idouglid, Lahcen; Tkatek, Said; Elfayq, Khalid 3 of 3

Abstract

The rapid evolution of cyber threats and the increasing complexity of networkconnected devices have expanded the need for robust Intrusion Detection Systems (IDS). This paper evaluates the effectiveness of five deep learning models--MLP, LSTM, CNN, RNN, and DNN--for host-based intrusion detection using the ADFA-LD dataset, this study compares models across metrics such as accuracy, precision, recall, and F1-score, highlighting the superiority of sequence-based models such as LSTM and RNN in detecting complex attack patterns. The results demonstrate that while RNN and LSTM provide the highest accuracy, they come with higher computational costs. The findings contribute to the growing field of IDS, offering insights into balancing performance and scalability in real-world applications. [ABSTRACT FROM AUTHOR]

Additional Information

  • Source:International Journal on Electrical Engineering & Informatics. 2025/03, Vol. 17, Issue 1, p63
  • Document Type:Article
  • Subject Area:Political Science
  • Publication Date:2025
  • ISSN:2085-6830
  • DOI:10.15676/ijeei.2025.17.1.5
  • Accession Number:185331362
  • Copyright Statement:Copyright of International Journal on Electrical Engineering & Informatics is the property of School of Electrical Engineering & Informatics, Bandung Institute of Technology and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Looking to go deeper into this topic? Look for more articles on EBSCOhost.