A Heuristic Intrusion Detection Approach Using Deep Learning Model
Publication Date
1-1-2023
Document Type
Conference Proceeding
Publication Title
2023 International Conference on Information Networking (ICOIN)
DOI
10.1109/ICOIN56518.2023.10049024
First Page
438
Last Page
442
Abstract
With the widespread usage of networking technology, Intrusion Detection System (IDS) attempts to identify and notify the users as normal or abnormal networking activities. In the wireless network systems, communication is via broadcast network packets. Black hat actors will attempt to compromise or cripple systems using wirelessly communicated packets. This paper proposes a heuristic approach to use a Deep Learning or Deep Neural Network (DNN) to evaluate the risk of intrusion from a given received network packet. The emphasis is on how a DNN can facilitate effective IDS with learning capability to accurately detect new or zero-day network behavior features and then rejecting the network intruder and reduce the risk of the network security. To demonstrate the effectiveness of the DNN model, we used CICIDS2018 dataset and support detection of eight behavioral issues in a network. The result of our IDS system achieved an accurate detection rate of 97% using 80% of the data.
Keywords
deep learning, DNN, IDS, neural net, web security
Department
Computer Science
Recommended Citation
Ching Seh Wu and Sam Chen. "A Heuristic Intrusion Detection Approach Using Deep Learning Model" 2023 International Conference on Information Networking (ICOIN) (2023): 438-442. https://doi.org/10.1109/ICOIN56518.2023.10049024
