Related references
Note: Only part of the references are listed.
Article
Computer Science, Information Systems
Wei Wang et al.
Summary: Network intrusion detection system is an essential cyber defense tool to protect systems from illegal attacks, and building an effective system using deep learning methods is challenging. The proposed RL-NIDS models network behavior by learning explicit and implicit feature interactions, achieving superior performance in multiclass classification compared to other methods.
COMPUTERS & SECURITY
(2022)
Article
Computer Science, Information Systems
Ayush Kumar et al.
Summary: In this work, a solution called EDIMA is presented for IoT botnet detection. It consists of lightweight modules deployed at edge gateways in home networks, with additional modules implemented on cloud servers. EDIMA aims to detect IoT botnets prior to an attack, featuring a novel two-stage Machine Learning-based detector. The solution achieves high accuracies in bot scanning and bot-CnC traffic detection, with low false positive rates, and is shown to be robust to increasing IoT device connections. Performance analysis demonstrates low bot detection delays and RAM consumption. EDIMA outperforms existing detection techniques for bot scanning traffic and bot-CnC server communication.
COMPUTERS & SECURITY
(2022)
Article
Engineering, Electrical & Electronic
Tao Wu et al.
Summary: This paper proposes a network intrusion detection algorithm based on enhanced random forest and the SMOTE algorithm. By using a hybrid algorithm to increase the number of minority samples, a balanced dataset is obtained. The prediction results are corrected using the similarity matrix of network attacks. Experimental results show an improvement in the classification accuracy of network intrusion detection.
EURASIP JOURNAL ON ADVANCES IN SIGNAL PROCESSING
(2022)
Article
Energy & Fuels
Yong Sun et al.
Summary: This paper presents a novel network anomaly detection framework based on data balance and feature selection, addressing the multiple classification of network intrusion. By introducing a resampling strategy and conducting feature selection based on information gain rate, the optimal feature selection scheme is obtained.
Article
Chemistry, Analytical
Jingmin Yang et al.
Summary: This paper addresses the issue of fluctuating received signal strength indication (RSSI) in Wi-Fi, which degrades the positioning accuracy. It proposes a calibration-free solution called signal strength difference (DIFF) for handling signal strength variance between devices. Additionally, a data hierarchical processing strategy and three positioning algorithms based on deep neural networks (DNN) are designed to extend indoor positioning to three dimensions. Experimental results demonstrate that the proposed algorithms outperform traditional machine learning and advanced deep learning algorithms in terms of positioning efficiency and accuracy.
Article
Multidisciplinary Sciences
Kezhou Ren et al.
Summary: This paper presents a network intrusion detection model based on RFE feature extraction and deep reinforcement learning. The model improves the efficacy of intrusion detection systems through feature selection and deep reinforcement learning, and demonstrates good performance in experiments.
SCIENTIFIC REPORTS
(2022)
Article
Mathematical & Computational Biology
S. K. B. Sangeetha et al.
Summary: The Internet of Things (IoT) connects a large array of objects, and securing the transport layer of IoT networks using a multilayered security approach based on deep learning has proven to be effective and achieved a high accuracy rate.
COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE
(2022)
Article
Mathematical & Computational Biology
Yue Li et al.
Summary: This paper proposes a hybrid intrusion detection method that utilizes ADASYN and ID3 decision tree to improve the effectiveness of intrusion detection rate. The model based on ADASYN and ID3 decision tree achieves higher accuracy and lower false alarm rate, making it more suitable for intrusion detection tasks.
MATHEMATICAL BIOSCIENCES AND ENGINEERING
(2022)
Review
Computer Science, Information Systems
Stavroula Bourou et al.
Summary: Recent technological innovations and the abundance of available data have led to an increase in cyberattacks against network systems. Intrusion Detection Systems (IDS) are crucial in defending networks against adversarial attacks, with machine learning methods providing various cybersecurity tools. The Generative Adversarial Network (GAN) is recognized for its potential in tabular data synthesis, and certain GANs have been evaluated as prominent in synthesizing network data.
Article
Physics, Multidisciplinary
Kun Zhou et al.
Summary: This research proposed an adaptable reinforcement learning model for anomaly detection, showing higher rewards and lower loss rates compared to classical machine learning and generative adversarial models. The experiment results demonstrated that the model achieved comparable or even higher precision, recall rate, and F1 score during training and testing processes.
Article
Computer Science, Information Systems
Wenfeng Xu et al.
Summary: This study proposed a novel intrusion detection system model (IDS-D-2) based on model-based interpretability, which combines normal and attack samples to highlight features and uses Additive Tree as a binary classifier to achieve good detection performance.
SECURITY AND COMMUNICATION NETWORKS
(2021)
Review
Chemistry, Analytical
Martin Pech et al.
Summary: With the advancement of new technologies in smart factories, automated predictive maintenance is becoming more closely connected to production robotization. Intelligent sensors enable easier data collection, but efficient analysis is crucial to support decision-making and management of complex systems. The review of current literature on predictive maintenance and intelligent sensors in smart factories provides an overview of future research challenges.
Article
Chemistry, Multidisciplinary
Hsiao-Chung Lin et al.
Summary: This study proposes an ensemble learning scheme based on a revised random forest algorithm for a security monitoring system in the domain of renewable energy to categorize network threats in a network intrusion detection system. The synthetic minority oversampling technique scheme (SMOTE) was formulated to re-balance the original data sets by altering the number of data points for minority class to reduce classification error for minority classes of experimental data in model training. The proposed classifier's classification performance in threat classification was experimentally verified in terms of accuracy, precision, recall, and F1-score on the UNSW-NB15 and CSE-CIC-IDS 2018 data sets.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Artificial Intelligence
Erfan A. Shams et al.
Summary: A new context-aware feature extraction method was proposed for CNN-based multiclass intrusion detection, which effectively improved classification accuracy by reducing feature space and classification time. The study showed that the method performed well on multiple datasets and enhanced the performance of intrusion detection.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Khairul Akram Zainol Ariffin et al.
Summary: The introduction of Industrial Revolution 4.0 has brought benefits but also raised concerns about cybercrimes. Digital forensic organizations must be prepared to handle large-scale cyberattacks in this era, and a systematic literature review has derived five indicators for maturity and readiness of DF organizations.
COMPUTERS & SECURITY
(2021)
Article
Automation & Control Systems
Beibei Li et al.
Summary: The study introduces a novel federated deep learning scheme named DeepFed for detecting cyber threats against industrial CPSs. By designing a new intrusion detection model and federated learning framework, the research successfully achieves secure detection of various cyber threats.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Computer Science, Artificial Intelligence
Vitali Herrera-Semenets et al.
Summary: The research proposed a novel feature selection algorithm for intrusion detection scenarios, which reduces the dimensionality of the training data set by using qualitative information provided by multiple feature selection measures, achieving greater efficacy than other feature selection algorithms for intrusion detection purposes. Future research should continue to improve the algorithm.
KNOWLEDGE-BASED SYSTEMS
(2021)
Review
Computer Science, Information Systems
Ayesha S. Dina et al.
Summary: Intrusions in computer networks have been on the rise in the past decade, leading researchers to propose signature-based and anomaly-based intrusion detection methods, with Machine Learning techniques playing a key role. This paper provides a comprehensive critical survey of ML-based intrusion detection approaches in the literature over the last ten years, highlighting some open issues for future research.
INTERNET OF THINGS
(2021)
Article
Computer Science, Theory & Methods
Achmad Akbar Megantara et al.
Summary: The internet has provided opportunities for innovation across various sectors but also poses risks of cyber-attacks. This research proposes a hybrid machine learning method to detect intrusion behaviors in network systems, showing high accuracy and stability in detecting certain attack types.
JOURNAL OF BIG DATA
(2021)
Article
Computer Science, Theory & Methods
Sugandh Seth et al.
Summary: This paper proposes a novel approach for a time-efficient and smart Intrusion Detection System by using a hybrid feature selection method to reduce prediction latency without affecting attack prediction performance. The proposed model with hybrid feature selection and LightGBM achieved high accuracy, sensitivity, and precision rates, as well as low prediction latency.
JOURNAL OF BIG DATA
(2021)
Article
Engineering, Mechanical
Junaid Arshad et al.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2020)
Article
Computer Science, Information Systems
Sweta Bhattacharya et al.
Article
Computer Science, Hardware & Architecture
Hongpo Zhang et al.
Article
Multidisciplinary Sciences
Chaofei Tang et al.
Article
Computer Science, Hardware & Architecture
Kamran Siddique et al.
Article
Chemistry, Analytical
Xiaopeng Tan et al.
Article
Chemistry, Multidisciplinary
Yanqing Yang et al.
APPLIED SCIENCES-BASEL
(2019)
