Related Experiment Video
Updated: Sep 8, 2025
03:31
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
Published on: December 15, 2023
633
A Network Intrusion Detection System Using Hybrid Multilayer Deep Learning Model
Muhammad Basit Umair1, Zeshan Iqbal1, Muhammad Ahmad Faraz2
1Department of Computer Science, University of Engineering and Technology Taxila, Taxila, Pakistan.
Big Data
|June 15, 2022
Summary
This study introduces a statistical approach for intrusion detection systems (IDS) to improve accuracy with large datasets. The proposed method achieved 99% accuracy, outperforming traditional methods in network security.
Area of Science:
- Computer Science
- Cybersecurity
- Network Security
Background:
- Traditional intrusion detection systems (IDS) struggle with high accuracy due to large data volumes.
- Existing methods often fail to effectively analyze complex network traffic patterns.
Purpose of the Study:
- To propose a novel statistical approach for enhancing intrusion detection system accuracy.
- To address the limitations of traditional methods in handling large-scale network data.
Main Methods:
- Feature extraction and selection using a multilayer convolutional neural network.
- Classification of network intrusions via a softmax classifier and a multilayer deep neural network.
- Validation using NSL-KDD and KDDCUP'99 benchmark datasets.
Main Results:
- The proposed statistical approach achieved a high accuracy of 99%.
- Performance metrics including recall, F1-score, and precision were used for evaluation.
- Superior performance demonstrated compared to existing intrusion detection systems.
Conclusions:
- The statistical approach offers a significant improvement for intrusion detection.
- The model effectively classifies network intrusions with high accuracy.
- This research provides a robust solution for modern network security challenges.
