Abstract: This project presents an innovative intrusion detection approach that utilizes the combined capabilities of deep neural networks (DNN), multilayer perceptron (MLP), long short-term memory (LSTM), convolutional neural networks (CNN), and fuzzy logic within a unified deep neuro- fuzzy framework. The key differentiator of this system lies in its strategic integration with Principal Component Analysis (PCA) during the training phase, which aims to increase feature representation and overall model performance. A distinctive feature of this system is the incorporation of PCA, a critical pre-processing step that plays a key role in extracting significant features from the CICIDS2017 and CICIDS2019 dataset. By using PCA, the dimensionality of the data set is substantially reduced, allowing the system to focus on the essential information necessary for effective intrusion detection. This dimensionality reduction leads to a remarkable improvement in feature representation, resulting in excellent model performance. PCA integration acts as a catalyst in the training phase, facilitating the extraction of relevant information and optimizing the deep neuro-fuzzy system for increased accuracy and robust generalization capabilities. The results show that this innovative approach not only improves the accuracy of intrusion detection, but also improves the system's ability to adapt to diverse and dynamic threats. Overall, the strategic use of PCA coupled with a unified deep neuro-fuzzy framework sets this system apart, making it a promising advancement in intrusion detection.
Keywords: Intrusion detection, Deep neural networks (DNN), Principal Component Analysis (PCA) Deep neuro-fuzzy framework
Cite:
Vignesh A Palan, Rathika Ramesh Gaunskar, Prashanth, Pruthviraj, Dr. Amirthavalli.M, "DEEP NEURAL FUZZY SYSTEM FOR INTRUSION DETECTION", IJARCCE International Journal of Advanced Research in Computer and Communication Engineering, vol. 13, no. 3, 2024, Crossref https://doi.org/10.17148/IJARCCE.2024.13333.
|
DOI:
10.17148/IJARCCE.2024.13333
