Enhancing Intrusion Detection Systems with Recurrent Neural Networks in D23 Deep Learning

DOI

https://www.doi.org/10.59256/ijire.20240503026

Published In

Volume 5, Issue 3 (2024)

Published Online

May-June 2024

Pages

209-216

Article Type

Research Article

Check for updates

Abstract: Strong cybersecurity safeguards are increasingly necessary as the digital landscape changes to protect sensitive data and vital infrastructure. To detect and reduce any risks to network security, intrusion detection systems, or IDS, are essential. To improve the precision and effectiveness of anomaly detection, this paper suggests a novel method of intrusion detection that makes use of recurrent neural networks (RNNs). Standard intrusion detection systems (IDS) frequently depend on static rule-based systems, which may find it difficult to adjust to changing and dynamic cyber threats. On the other hand, sequence-retention neural networks, or RNNs, they have a special capacity to recognize patterns and temporal connections in sequential input. Traditional IDS methods are contrasted with the suggested system, which is assessed using common datasets. The RNN-based IDS is successful at detecting known and unknown threats while reducing false positives, according to preliminary data.