Use of Singular Value Decomposition for a Deep Learning-Based Fast Intrusion Detection System

Abstract

An artificial intelligence technique based on artificial neural networks for identifying risks. A range of professions, including the recognition of specific patterns or categories, have adopted deep learning methodologies. Data from intrusion detection assessments and security event monitoring were used to evaluate the network situation. The performance and accuracy of the detection must be improved. We decided to test a range of approaches utilizing an open data set in order to identify the best approach for intrusion detection. The current study aims to explore the possibility of using singular value decomposition (SVD) as a pre-processing step to reduce the dimensionality of the data. In addition to reducing the noise from the data, this pre-processing step reduces the dimensionality of the data to save time on calculations. The proposed strategy can help other currently used methods perform better. We test reduction strategies on the UNSW-NB15 dataset, and the outcomes are very positive.