Abstract:A large number of network security breaches in IoT networks have demonstrated the unreliability of current Network Intrusion Detection Systems (NIDSs). Consequently, network interruptions and loss of sensitive data have occurred, which led to an active research area for improving NIDS technologies. In an analysis of related works, it was observed that most researchers aim to obtain better classification results by using a set of untried combinations of Feature Reduction (FR) and Machine Learning (ML) techniques on NIDS datasets. However, these datasets are different in feature sets, attack types, and network design. Therefore, this paper aims to discover whether these techniques can be generalised across various datasets. Six ML models are utilised: a Deep Feed Forward (DFF), Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), Decision Tree (DT), Logistic Regression (LR), and Naive Bayes (NB). The accuracy of three Feature Extraction (FE) algorithms; Principal Component Analysis (PCA), Auto-encoder (AE), and Linear Discriminant Analysis (LDA), are evaluated using three benchmark datasets: UNSW-NB15, ToN-IoT and CSE-CIC-IDS2018. Although PCA and AE algorithms have been widely used, the determination of their optimal number of extracted dimensions has been overlooked. The results indicate that no clear FE method or ML model can achieve the best scores for all datasets. The optimal number of extracted dimensions has been identified for each dataset, and LDA degrades the performance of the ML models on two datasets. The variance is used to analyse the extracted dimensions of LDA and PCA. Finally, this paper concludes that the choice of datasets significantly alters the performance of the applied techniques. We believe that a universal (benchmark) feature set is needed to facilitate further advancement and progress of research in this field.

Effective Feature Extraction for Intrusion Detection System using Non-negative Matrix Factorization and Univariate analysis

A Mixed Intrusion Detection System Utilizing K-means and Extreme Gradient Boosting

A Deep Learning-Based Framework for Feature Extraction and Classification of Intrusion Detection in Networks

An Integrated Feature Extraction Based on Principal Components and Deep Auto Encoder with Extra Tree for Intrusion Detection Systems

A computationally efficient dimensionality reduction and attack classification approach for network intrusion detection

Feature Extraction for Machine Learning-based Intrusion Detection in IoT Networks

Performance Analysis of Intrusion Detection Systems Using a Feature Selection Method on the UNSW-NB15 Dataset

An Explainable Intrusion Detection System Based on Feature Importance

A Hybrid Modified Deep Learning Architecture for Intrusion Detection System with Optimal Feature Selection

Intrusion detection system over real-time data traffic using machine learning methods with feature selection approaches

An Intrusion Detection Model Based on Feature Reduction and Convolutional Neural Networks

Relevant Feature Selection Model Using Data Mining for Intrusion Detection System

2-layer classification model with correlated common feature selection for intrusion detection system in networks

Implement an Intrusion Detection System Utilizing Machine Learning and Principal Component Analysis

Performance Analysis of Anomaly-Based Network Intrusion Detection Using Feature Selection and Machine Learning Techniques

DL-IDS: Extracting Features Using CNN-LSTM Hybrid Network for Intrusion Detection System

Fast intrusion detection based on a non-negative matrix factorization model

Analysis of Different Dimensionality Reduction Techniques and Machine Learning Algorithms for an Intrusion Detection System

Application and Performance Analysis of Data Preprocessing for Intrusion Detection System.

Fusion of statistical importance for feature selection in Deep Neural Network-based Intrusion Detection System

Intrusion detection systems using classical machine learning techniques versus integrated unsupervised feature learning and deep neural network