Article
Computer Science, Information Systems
K. Narayana Rao, K. Venkata Rao, P. V. G. D. Prasad Reddy
Summary: The study found that machine learning has shown good results in intrusion detection systems. The two-stage hybrid methodology proposed by the authors significantly improves the detection of attacks, especially achieving excellent accuracy and detection rates on the UNSW-NB15 dataset.
COMPUTER COMMUNICATIONS
(2021)
Article
Biochemistry & Molecular Biology
Abdullah Al Mamun, Raihanul Bari Tanvir, Masrur Sobhan, Kalai Mathee, Giri Narasimhan, Gregory E. Holt, Ananda Mohan Mondal
Summary: The study utilized deep learning algorithms CAE and mrCAE to identify key lncRNAs for differentiating the origin of various cancers, with mrCAE outperforming single-run CAE, AE, and other feature selection techniques. They identified a set of top 128 lncRNAs that could accurately differentiate the origin of 12 different cancers.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2021)
Article
Computer Science, Information Systems
Ivandro O. Lopes, Deqing Zou, Ihsan H. Abdulqadder, Francis A. Ruambo, Bin Yuan, Hai Jin
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
Abdullah I. A. Alzahrani, Amal Al-Rasheed, Amel Ksibi, Manel Ayadi, Mashael M. M. Asiri, Mohammed Zakariah
Summary: The translation discusses the importance of securing sensitive data in Internet of Things devices. It introduces a new intrusion detection model deployed at fog nodes to detect undesired traffic towards IoT devices. The Tab transformer model is proposed and shows high accuracy in classifying normal and abnormal traffic data as well as predicting multiple class attacks.
Article
Computer Science, Artificial Intelligence
Qigang Liu, Deming Wang, Yuhang Jia, Suyuan Luo, Chongren Wang
Summary: With the frequent occurrence of cyber-security incidents, intrusion detection system (IDS) has been given more attention, but accurately detecting attacks from traffic data stream is challenging due to the diverse nature of network intrusions and serious imbalanced class distribution. Traditional methods suffer from drawbacks, while the proposed models for intrusion detection outperform state-of-art methods on both binary and multi-class classification, by addressing the issues of feature extraction difficulties and imbalanced data distribution through customized loss function.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Energy & Fuels
Ruizhe Yao, Ning Wang, Zhihui Liu, Peng Chen, Di Ma, Xianjun Sheng
Summary: This paper proposes a feature engineering based AutoEncoder (AE)-LightGBM intrusion detection system for SDN, which improves intrusion detection performance by optimizing data distribution and feature extraction. Experimental results show better accuracy, precision, and F1-score performance compared to traditional models and related works.
Article
Computer Science, Information Systems
Danijela Protic, Miomir Stankovic, Radomir Prodanovic, Ivan Vulic, Goran M. Stojanovic, Mitar Simic, Gordana Ostojic, Stevan Stankovski
Summary: Anomaly-based intrusion detection systems classify computer network behavior by identifying deviations from the statistical model of typical behavior. Feature selection and feature scaling are commonly used techniques to improve classifier performance.
Article
Computer Science, Information Systems
Giuseppina Andresini, Annalisa Appice, Donato Malerba
Summary: In this study, a new intrusion detection method is introduced which leverages a deep metric learning methodology combining autoencoders and Triplet networks. Two separate autoencoders are trained on historical normal network flows and attacks, and a Triplet network is trained to learn the embedding of the feature vector representation of network flows. This methodology achieves better predictive accuracy in detecting new signs of malicious activities in network traffic compared to competitive intrusion detection architectures on benchmark datasets.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Information Systems
Manar Ahmed Hamza, Aisha Hassan Abdalla Hashim, Heba G. Mohamed, Saud S. Alotaibi, Hany Mahgoub, Amal S. Mehanna, Abdelwahed Motwakel
Summary: Internet of Everything (IoE) is an interconnected network of people, processes, data, and things. Security and privacy challenges in IoE environment can be addressed by developing effective Intrusion Detection Systems (IDS). This study proposes an Intelligent Multiverse Optimization with Deep Learning Enabled Intrusion Detection System (IMVO-DLIDS) for IoT environment, which focuses on identification and classification of intrusions. The proposed model utilizes data pre-processing, feature selection, and classification algorithms to achieve promising outcomes compared to other approaches.
CMC-COMPUTERS MATERIALS & CONTINUA
(2022)
Article
Computer Science, Hardware & Architecture
Juan-juan Fu, Xing-lan Zhang
Summary: This paper proposes a feature fusion technique based on gradient importance enhancement to improve the accuracy and generalization ability of the intrusion detection model in the current unstable network security situation.
Article
Neurosciences
Huaiqiang Sun, Guoting Luo, Su Lui, Xiaoqi Huang, John Sweeney, Qiyong Gong
Summary: This study proposes a specially designed autoencoder to investigate structural brain changes in schizophrenia. The classifier trained with autoencoded features outperforms the classifier trained with conventional morphological features in identifying schizophrenia patients from healthy controls.
HUMAN BRAIN MAPPING
(2023)
Article
Chemistry, Analytical
Naoto Yoshimura, Hiroki Kuzuno, Yoshiaki Shiraishi, Masakatu Morii
Summary: This paper proposes a new intrusion detection system model called DOC-IDS, based on Perera's deep one-class classification. The model uses three different loss functions for training and utilizes open datasets for feature extraction. Experimental results show that the DOC-IDS offers improved anomaly detection performance while reducing the load resulting from the design and extraction of feature values.
Article
Computer Science, Information Systems
Lianming Zhang, Kui Liu, Xiaowei Xie, Wenji Bai, Baolin Wu, Pingping Dong
Summary: The paper proposes a data-driven NIDS based on feature selection and deep learning, called FS-DL. FS-DL improves detection accuracy by enhancing data quality and reducing computational load. Experimental results demonstrate that FS-DL achieves better detection performance with only a small number of features, and it has been deployed in an SDN controller for online detection of abnormal traffic.
JOURNAL OF INFORMATION SECURITY AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Shubhra Dwivedi, Manu Vardhan, Sarsij Tripathi
Summary: The EFSGOA method, a combination of ensemble feature selection and grasshopper optimization algorithm, achieved excellent performance in intrusion detection, with high detection rates, accuracy, and low false alarm rates. The method significantly improved accuracy and reduced false alarms, outperforming other existing techniques.
CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Sydney Mambwe Kasongo
Summary: In recent years, advances in technologies such as cloud computing, vehicular networks systems, and the Internet of Things (IoT) have led to a spike in the amount of information transmitted through communication infrastructures. Consequently, attackers have increased their efforts to exploit vulnerabilities in network systems. Therefore, it is crucial to enhance the security of these network systems. This study implements an IDS framework using Machine Learning techniques and evaluates its performance using benchmark datasets.
COMPUTER COMMUNICATIONS
(2023)