Related references
Note: Only part of the references are listed.
Article
Automation & Control Systems
Mohammed M. Alani et al.
Summary: The Internet of Things (IoT) has become a driving paradigm in various applications, but its security vulnerabilities and threats have negative impacts on deployment and operation. This article presents an intelligent two-layer intrusion detection system for IoT, using machine learning techniques to handle flow and packet features and minimizing time overhead by selecting significant features for accurate intrusion detection.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Chemistry, Analytical
Asaad Balla et al.
Summary: Integrating IoT devices in SCADA systems has improved data collection and transmission technologies, but also presents security challenges. Many studies on IDS in SCADA systems neglect the fundamental issue of imbalanced datasets. This research investigates the impact of data imbalance on the effectiveness of a SCADA-based IDS.
Article
Computer Science, Artificial Intelligence
Dheeb Albashish et al.
Summary: This paper proposes a heterogeneous ensemble classifier configuration for a multiclass intrusion detection problem. The ensemble is composed of k-nearest neighbors, artificial neural networks, and naive Bayes classifiers, and the decisions of these classifiers are combined with weighted majority voting. The empirical study shows that the ensemble configuration using ACOR-optimized weights is capable of resolving conflicts between multiple classifiers and improving classification accuracy.
Article
Telecommunications
C. U. Om Kumar et al.
Summary: In communication and information technology, the Internet of Things (IoT) generates significant data traffic that can be analyzed to identify unusual network load and trends. To overcome limitations in existing intrusion detection models (IDM), a DL-based RKCNN-MMBO model is proposed for attack identification. Pre-processing involves min_max normalization and feature selection using the IBRO algorithm, with classification performed by the RKCNN classifier. The RKCNN-MMBO classifier achieves higher accuracy in IDM detection compared to other methods.
WIRELESS PERSONAL COMMUNICATIONS
(2023)
Article
Computer Science, Information Systems
Ghulam Muhammad et al.
Summary: With the rapid growth of the Internet of Things (IoT) and the increase in data volume and network traffic, it has become easier for intruders to launch network attacks. This article proposes an intrusion detection system (IDS) based on stacked autoencoders (AE) and deep neural networks (DNN) to address this issue. The system achieved high accuracy rates of 94.2%, 99.7%, and 99.9% for multiclass classification on different datasets.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Automation & Control Systems
Tej Kiran Boppana et al.
Summary: Households and businesses are increasingly reliant on IoT systems, leading to a steep increase in the number of IoT devices. Manufacturers are focused on producing cheaper and easier-to-use devices, neglecting the importance of security. Novel attacks are targeting supervised intrusion detection models used to protect IoT systems. To address this, a novel unsupervised GAN and autoencoder-based model called GAN-AE is proposed for detecting unknown intrusions in MQTT IoT applications.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Hardware & Architecture
Zecheng Li et al.
Summary: The article proposes an accurate abnormal traffic detection method using pseudoanomaly for semisupervised detection. It utilizes an efficient feature extraction framework and a novel denoise autoencoder-generative adversarial network model to extract spatial and temporal features from traffic and generate high-quality pseudoanomalies. Experimental results show that the method achieves high precision and recall on different datasets.
IEEE TRANSACTIONS ON RELIABILITY
(2023)
Article
Computer Science, Hardware & Architecture
T. Lavanya et al.
Summary: Semiconductor technologies have gained wide attention across industries, but concerns about intrusion, particularly Hardware Trojans, have been raised due to high design-complexities and untrusted foundries. This paper proposes a machine learning-driven Hardware Trojan detection system using a robust register transfer logic (RTL) analysis assisted by multi-parameter approach. The proposed model achieved nearly 100% accuracy, F-Measure, and Area Under Curve (AUC) for path-level Trojan detection, outperforming known approaches.
INTEGRATION-THE VLSI JOURNAL
(2023)
Article
Computer Science, Hardware & Architecture
Murad Ali Khan et al.
Summary: Traditional ML based IDS cannot handle high-speed and ever-evolving attacks. This study proposes an OE-IDS model using AutoML based on a soft voting method for detecting intrusion in the network environment. The proposed model achieves high accuracy and minimizes false alarm rates through optimal ensemble strategy and different sampling methods.
JOURNAL OF NETWORK AND COMPUTER APPLICATIONS
(2023)
Article
Chemistry, Analytical
Yung-Chung Wang et al.
Summary: This study utilizes the latest CSE-CIC-IDS2018 dataset and evaluation metrics to construct six models for detecting malicious attacks in network traffic. Each model achieves high accuracy and the multi-class classification accuracy is above 98%. Compared to other intrusion detection systems, the proposed model demonstrates improved performance.
Article
Computer Science, Information Systems
Sarhad Arisdakessian et al.
Summary: This article presents an in-depth survey on existing intrusion detection solutions for IoT ecosystem, including the communications between IoT devices, fog computing, and cloud computing layers. It proposes a novel two-level classification scheme and a comprehensive cybersecurity framework that combines concepts of explainable artificial intelligence (XAI), federated learning, game theory, and social psychology to provide strong protection against cyberattacks for future IoT systems.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Computer Science, Information Systems
Yanze Qu et al.
Summary: In this paper, a network intrusion detection method based on domain confusion is proposed to improve the migration performance of deep learning models. It utilizes a domain confusion network for feature transformation, mapping traffic data in different network environments to the same feature space. Experiment results demonstrate that the proposed method achieves comparable or even better detection performance compared to traditional models, and shows better migration performance in different network environments.
Article
Computer Science, Information Systems
Chao Wang et al.
Summary: In this study, a novel semi-supervised anomaly detection-based IDS is proposed, leveraging the capabilities of representation learning and two anomaly detectors. Firstly, an autoencoder (AE) is used to extract representative features of normal data, followed by training two semi-supervised detectors, namely one-class support vector machine (OCSVM) and Gaussian mixture model (GMM), on the derived features. The two detectors collaborate to detect anomalous samples, with OCSVM predicting abnormal samples initially and GMM rechecking the misclassified samples further. Experiments demonstrate that AE improves the detection rate, and two detectors are more promising than a single one.
Article
Computer Science, Information Systems
Taehoon Kim et al.
Summary: The currently used network-intrusion detection systems (NIDSs) based on deep learning face limitations in processing large amounts of data in real time due to time-consuming flow information collection and feature creation. This study proposes a novel NIDS model with ?(1) memory complexity for flow processing, which can handle numerous concurrent flows thanks to its small memory requirement. The proposed NIDS uses raw packet data as input features for deep learning models, resulting in a lightweight feature-creation process. A validation model is also introduced to address false detection, leading to high detection accuracy. Real-time detection is achievable through intrusion detection on every received packet using the Inception model. Performance comparison shows significantly improved detection time and reduced memory requirement by 73% and 77% on average, while maintaining high detection accuracy. Therefore, the proposed model effectively tackles the challenges of modern deep-learning-based NIDSs.
Article
Physics, Applied
Ammar Alazab et al.
Article
Computer Science, Information Systems
Minxiao Wang et al.
Summary: Machine learning-based Network Intrusion Detection Systems (NIDSs) can classify network flow behavior as benign or malicious. However, current ML-based NIDSs are insufficient to generalize in changing network environments like IoT-based smart homes. This paper proposes a novel Transformer-based IoT NIDS method that combines network traffic-based and telemetry data-based NIDSs to achieve improved intrusion detection performance.
Article
Computer Science, Artificial Intelligence
Jin Xie et al.
Summary: In this paper, two algorithms based on the random vector functional link network (RVFLN) and alternating direction method of multipliers algorithm are proposed to solve distributed learning (DL) problems with datasets containing outliers. The algorithms work in a fully distributed manner and are privacy-preserving methods. The experiments show that the proposed algorithms are robust and efficient in distributed learning with data including outliers, and the Huber-DRVFL algorithm is more stable than the L-1-DRVFL algorithm when the parameters vary.
ARTIFICIAL INTELLIGENCE REVIEW
(2023)
Article
Computer Science, Information Systems
Vikash Kumar et al.
Summary: This paper proposes a method that uses Wasserstein Conditional Generative Adversarial Network (WCGAN) and XGBoost classifier to accurately detect a large number of attack classes and solve the problem of imbalanced data samples. Experimental results show that this method achieves significant improvement in performance metrics.
COMPUTERS & SECURITY
(2023)
Article
Chemistry, Analytical
Shu-Ming Tseng et al.
Summary: This paper introduces a transformer encoding linker network (TELNet) that automatically identifies scene boundaries in videos without prior knowledge of their structure. TELNet efficiently determines scene boundaries by encoding and linking video shots using a rolling window. TELNet demonstrated comparable results to other state-of-the-art models in standard settings and significantly improved results in cross-dataset evaluations.
Article
Computer Science, Theory & Methods
Md Alamgir Hossain et al.
Summary: Intrusion detection is crucial in network security for protecting computer systems. Traditional signature-based intrusion detection systems have limitations in identifying unknown threats. This research proposes a novel ensemble-based machine-learning technique for intrusion detection and evaluates its performance using various public datasets and ensemble strategies.
Proceedings Paper
Computer Science, Artificial Intelligence
Md Mohaiminul Islam et al.
Summary: The ability to accurately detect movie scenes is crucial for understanding the storyline, but it is often challenging. This study proposes TranS4mer, a novel model that efficiently captures dependencies in long movie videos for accurate scene detection. TranS4mer outperforms prior methods in multiple datasets and is 2x faster with 3x less GPU memory usage compared to standard Transformer models.
2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2023)
Article
Computer Science, Hardware & Architecture
Felix Erlacher et al.
Summary: Signature-based Network Intrusion Detection Systems (NIDS) are the leading technology in modern network security solutions, but they often struggle to handle high-speed network links. Flow-monitoring based on the IPFIX standard provides an alternative solution for high-speed networks, allowing for collection of statistical information and inclusion of payload in flows to be used with NIDS. This paper presents an improved version of the IPFIX-based Signature-based Intrusion Detection System (FIXIDS), which can handle higher data rates without drops compared to Snort while maintaining the same event detection rate.
IEEE TRANSACTIONS ON DEPENDABLE AND SECURE COMPUTING
(2022)
Article
Computer Science, Theory & Methods
Randhir Kumar et al.
Summary: The Internet of Things (IoT) is a new technology for critical application development, but faces challenges in privacy, security, and single point of failure. Blockchain technology can address these challenges and integrating it with IoT can benefit individuals and society. This paper proposes a distributed Intrusion Detection System (IDS) using fog computing to detect DDoS attacks in blockchain-enabled IoT networks, and evaluates the performance using machine learning algorithms.
JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING
(2022)
Article
Computer Science, Information Systems
Manisha Rani et al.
Summary: The Intrusion Detection System is crucial for network security and the proposed deep neural network effectively addresses the problem of class imbalance to improve classification accuracy. The experiments demonstrate the superiority of the approach in detecting malicious activities accurately.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Li Yang et al.
Summary: Modern vehicles, including connected vehicles and autonomous vehicles, are vulnerable to cyber-attacks due to their increasing functionality and connectivity. To secure vehicular networks, researchers propose a hybrid intrusion detection system that can detect both known and unknown attacks.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Computer Science, Artificial Intelligence
Earum Mushtaq et al.
Summary: This study proposes a hybrid framework for intrusion detection system using a combination of deep auto-encoder (AE) and long short term memory (LSTM). Experimental results show that the proposed AE-LSTM model achieves higher classification accuracy and detection rate on the NSL-KDD dataset compared to other deep and shallow machine learning techniques, indicating its superior performance.
APPLIED SOFT COMPUTING
(2022)
Article
Computer Science, Information Systems
Mohamed Hammad et al.
Summary: This paper proposes a novel approach for the classification of network attacks, which is of great significance in network security. By analyzing the key factors in network traffic, reducing data dimension, and addressing the imbalance issue in the dataset, high detection accuracy and low false positive rate are achieved.
COMPUTERS & SECURITY
(2022)
Article
Computer Science, Hardware & Architecture
Chang Liu et al.
Summary: This article focuses on the network intrusion detection model based on data level and proposes three data processing schemes, which have been proven effective through experiments.
IEEE TRANSACTIONS ON RELIABILITY
(2022)
Article
Computer Science, Interdisciplinary Applications
Yanling Zhang et al.
Summary: This paper presents a hybrid intrusion detection framework, including the use of fuzzy association rules and the K-Means algorithm for normal mode and major attack detection, as well as the use of an active learning-based rare category detection algorithm to identify attack patterns. By integrating expert decisions in an interactive visualization system, the effectiveness and interpretability of the detection process are improved.
JOURNAL OF VISUALIZATION
(2022)
Article
Computer Science, Information Systems
Segun Popoola et al.
Summary: This article proposes a federated deep learning method for zero-day botnet attack detection in IoT-edge devices. The method generates a global model by coordinating the training of independent models in multiple IoT-edge devices, achieving high-performance classification detection of zero-day botnet attacks and ensuring data privacy and security.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Chemistry, Analytical
Danish Attique et al.
Summary: This study proposes a fog-assisted deep learning-based intrusion detection system for resource-constrained smart industries. The simulation results demonstrate the superior performance of the proposed system.
Article
Computer Science, Theory & Methods
Ying Zhang et al.
Summary: This article proposes an intrusion detection method for the IoT based on Improved Conditional Variational Autoencoder (ICVAE) and Borderline Synthetic Minority Oversampling Technique (BSM). Experimental results show that the proposed method can more effectively improve the accuracy of IoT attack detection under the condition of unbalanced samples.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2022)
Article
Computer Science, Artificial Intelligence
Xiangjun Li et al.
Summary: This paper proposes an intrusion detection algorithm based on unbalanced learning, which addresses the issues caused by unbalanced data through data grouping strategy and weighted loss function.
JOURNAL OF EXPERIMENTAL & THEORETICAL ARTIFICIAL INTELLIGENCE
(2022)
Article
Computer Science, Information Systems
Giovanni Apruzzese et al.
Summary: Enhancing Network Intrusion Detection Systems (NIDS) with supervised Machine Learning (ML) is difficult due to the need for labeled data. We propose using existing labeled data for cross-evaluations of ML-NIDS to discover unknown qualities. We introduce the first cross-evaluation model and framework, demonstrating the potential and risks of cross-evaluations.
IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT
(2022)
Article
Computer Science, Information Systems
Ivandro O. Lopes et al.
Summary: This study introduces a semi-supervised intrusion detection framework that combines unsupervised and supervised techniques to address the lack of labeled network traffic. By using unsupervised pre-training and DNN classifier training, it achieves efficient intrusion detection and outperforms other competitive methods.
COMPUTER COMMUNICATIONS
(2022)
Article
Computer Science, Information Systems
Chunying Zhang et al.
Summary: This paper systematically reviews the application literature of machine learning algorithms in network intrusion detection over the past decade, and compares the performance of different algorithms through experiments on various datasets. The results show that ensemble learning algorithms generally perform better, while Naive Bayes algorithm excels in dealing with new types of attacks. The performance of deep learning algorithms in this experiment is not particularly remarkable, but is influenced by factors such as structure and hyperparameters.
COMPUTERS & SECURITY
(2022)
Article
Computer Science, Artificial Intelligence
Jialiang Xie et al.
Summary: This article proposes a network intrusion-detection algorithm based on dynamic intuitionistic fuzzy sets, which can effectively detect and analyze abnormal network behavior. Experimental results show that the proposed algorithm outperforms other algorithms in terms of classification performance on classic network intrusion datasets.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2022)
Article
Computer Science, Information Systems
C. Prajisha et al.
Summary: This paper proposes an efficient intrusion detection mechanism for MQTT-IoT networks using an enhanced chaotic salp swarm optimization algorithm (ECSSA) and LightGBM classifier. Experimental results show that the proposed mechanism improves the overall accuracy rate and outperforms existing approaches in several test sets.
INTERNATIONAL JOURNAL OF INFORMATION SECURITY
(2022)
Article
Computer Science, Information Systems
Akbar Telikani et al.
Summary: Cyber attacks and intrusions pose significant barriers to the adoption of Industrial Internet of Things (HoT) in critical industries. Researchers have introduced the EvolCostDeep model to address the problem of imbalanced data distribution in HoT environments, and have designed the DeepIDSFog framework to improve the scalability of HoT IDSs.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Computer Science, Information Systems
Firuz Kamalov et al.
Summary: This paper proposes a novel ensemble classification method that deals with imbalanced data by training each tree in the ensemble using uniquely generated synthetically balanced data. Data balancing is achieved through kernel density estimation, resulting in a lower variance of the model estimator. The proposed classifier significantly outperforms benchmark methods in various datasets.
Article
Computer Science, Information Systems
P. L. S. Jayalaxmi et al.
Summary: The increasing number of connected devices in the era of IoT has led to an increase in intrusions. Intrusion Detection Systems (IDS) and Intrusion Prevention Systems (IPS) are important for developing security models, but there is a lack of coherence and advancements in existing models. Our survey presents a risk factor analysis and proposes a hybrid framework for an efficient security model for intrusion detection and/or prevention. We emphasize the importance of AI-based techniques, particularly Machine Learning and Deep Learning, for IoT security systems.
Proceedings Paper
Computer Science, Artificial Intelligence
Kumar Saurabh et al.
Summary: The proliferation of IoT devices in various sectors has led to a surge in botnet attacks, posing a threat to IoT security. To address this issue, a semi-supervised deep learning approach using SGAN for IoT botnet detection has been proposed and proven to be more accurate and efficient than existing supervised solutions.
2022 IEEE INTERNATIONAL CONFERENCE ON OMNI-LAYER INTELLIGENT SYSTEMS (IEEE COINS 2022)
(2022)
Article
Telecommunications
Qasem Abu Al-Haija et al.
Summary: In this paper, we propose an ensemble learning model called ELBA-IoT for botnet attack detection in IoT networks. The experimental results demonstrate that ELBA-IoT can detect botnet attacks with high accuracy and low overhead, outperforming other state-of-the-art techniques.
JOURNAL OF SENSOR AND ACTUATOR NETWORKS
(2022)
Article
Computer Science, Theory & Methods
Giuseppina Andresini et al.
Summary: This paper presents a deep learning methodology for binary classification of network traffic by representing network flows as 2D images and using Generative Adversarial Networks (GAN) and Convolutional Neural Networks (CNN). The GAN is trained to generate new images of unforeseen network attacks to augment training data for a CNN-based intrusion detection model, leading to better predictive accuracy.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(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
Xing Xu et al.
Summary: Intrusion detection is important for network security, but traditional machine learning-based methods face challenges with unknown attacks. This article introduces a novel deep learning method LCVAE for intrusion detection, which effectively handles imbalanced intrusion data. Experimental results demonstrate the superior performance of LCVAE in detecting intrusions and its potential for generating diverse new intrusion data.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Computer Science, Information Systems
Jingmei Liu et al.
Summary: The study proposes a network intrusion detection system based on ADASYN oversampling technology and LightGBM, which addresses the issue of imbalanced data by preprocessing and increasing minority samples to enhance detection accuracy.
COMPUTERS & SECURITY
(2021)
Article
Computer Science, Information Systems
Zhihan Lv et al.
Summary: The study constructed a hierarchical intrusion security detection model based on the autoencoder, achieving high accuracy and low false-negative and false-positive rates. Compared with other deep learning algorithms and common feature dimension reduction methods, the model demonstrated superior performance.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Review
Computer Science, Information Systems
Ansam Khraisat et al.
Summary: This paper provides a comprehensive review of contemporary IoT IDS, including techniques, deployment strategies, validation methods, and datasets for building IDS. It also discusses the classification of IoT attacks and future research challenges in countering these attacks.
Article
Computer Science, Information Systems
Kamaldeep et al.
Article
Computer Science, Information Systems
Georgios Douzas et al.
INFORMATION SCIENCES
(2019)
Article
Chemistry, Multidisciplinary
Yanqing Yang et al.
APPLIED SCIENCES-BASEL
(2019)
Article
Computer Science, Information Systems
Yair Meidan et al.
IEEE PERVASIVE COMPUTING
(2018)
