Article
Computer Science, Information Systems
ThankGod Obasi, M. Omair Shafiq
Summary: This study proposes an ensemble learning model for the classification of encrypted network traffic data, using machine learning and deep learning techniques to extract statistical features from encrypted network traffic. The experiment results show that the proposed model achieves high performance in terms of accuracy and AUC.
COMPUTER COMMUNICATIONS
(2022)
Article
Chemistry, Multidisciplinary
Sultan Almuhammadi, Abdullatif Alnajim, Mohammed Ayub
Summary: The QUIC protocol offers advantages over traditional TCP, but its encryption functionality reduces visibility into network traffic. This study utilizes five ensemble machine learning techniques to classify QUIC traffic and evaluates their performance using a publicly available dataset. The results show that Extreme Gradient Boosting Tree and Light Gradient Boosting Model outperform other models with simpler model architectures and features.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Information Systems
Yanlu Gong, Quanwang Wu, Mengchu Zhou, Junhao Wen
Summary: Multi-label learning aims to solve classification problems where instances are associated with a set of labels. This work proposes a novel approach called Self-paced Multi-label Co-Training (SMCT) that leverages the co-training paradigm to train two classifiers iteratively and communicate predictions on unlabeled data. Experimental evaluations demonstrate the competitive performance of SMCT compared to state-of-the-art methods.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Hardware & Architecture
Suguna Paramasivam, R. Leela Velusamy
Summary: According to a study, network traffic classification is a fundamental and complex part in software-defined networking. For 5G, end-to-end security is required and the network traffic is automatically classified using its software-defined architecture. To enhance the security in network traffic classification, a correct set of policy rules needs to be applied. Machine learning is used in the SDN environment to create an efficient and intelligent traffic classification framework.
JOURNAL OF SUPERCOMPUTING
(2023)
Article
Computer Science, Information Systems
Parvin Ahmadi Doval Amiri, Samuel Pierre
Summary: In this paper, an ensemble-based machine learning model is proposed for network traffic prediction in VANET. The proposed model combines different machine learning models using ensemble learning to achieve better performance and accuracy. Boruta and LightGBM are used as ensemble feature selection methods to consider the most informative attributes of the VANET dataset. The simulation results show that the proposed model outperforms the single algorithm in terms of prediction stability, accuracy, and execution time.
Article
Computer Science, Hardware & Architecture
Kunda Lin, Xiaolong Xu, Honghao Gao
Summary: The paper introduces a novel scheme for identifying encrypted traffic, TSCRNN, which automatically extracts features for efficient traffic classification. The results show superior performance in various classification tasks, indicating its potential to support the development of core technologies in the Industrial Internet of Things.
Article
Chemistry, Multidisciplinary
Fei Liu, Jiawei Li, Xiangxi Wen, Yu Wang, Rongjia Tong, Shubin Liu, Daxiong Chen
Summary: This paper introduces a model for assessing air traffic situation based on complex network theory and ensemble learning. By introducing an air traffic weighted network model and complex network analysis method, the paper proposes an objective and accurate evaluation criterion, and transforms the evaluation into a binary classification, which can accurately determine the air traffic situation.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Artificial Intelligence
Wasif Arman Haque, Samin Arefin, A. S. M. Shihavuddin, Muhammad Abul Hasan
Summary: The paper presents a novel energy-efficient Thin yet Deep convolutional neural network architecture for traffic sign recognition with less than 50 features in each convolutional layer, achieving superior performance on two publicly available datasets.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Jing Huang, Yang Peng, Lin Hu
Summary: This research aims to improve driving safety by recognizing the driver's mental load and emotional states. The proposed improved feature selection algorithm and multilayer stacking ensemble learning method have been validated to enhance the accuracy and reliability of driver state detection.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Article
Computer Science, Hardware & Architecture
Amin Shahraki, Mahmoud Abbasi, Amir Taherkordi, Anca Delia Jurcut
Summary: This paper investigates and compares online machine learning techniques for data stream analytics in the networking domain. The importance of traffic data analytics and the advantages of online learning are highlighted, along with the challenges associated with online learning-based network traffic stream analysis. The paper reviews data stream processing tools and frameworks, and evaluates the performance of different algorithms for network traffic classification. Finally, it presents open issues and future directions in analyzing traffic data streams.
Article
Computer Science, Information Systems
Ramazan Bozkir, Murtaza Cicioglu, Ali Calhan, Cengiz Togay
Summary: This study presents a new platform for classifying encrypted network traffic using machine learning techniques. The platform is designed to address real-world network traffic classification problems with performance-oriented, practical, and up-to-date software technologies. A new feature extraction method is also introduced in this study. The experimental results demonstrate that the XGBoost algorithm achieves the highest performance with an F1 score above 99%.
COMPUTER COMMUNICATIONS
(2023)
Article
Telecommunications
Amin Shahraki, Mahmoud Abbasi, Amir Taherkordi, Anca Delia Jurcut
Summary: This study investigates the applicability of Active Learning (AL) in Network Traffic Classification (NTC), discusses its challenges and open issues, and demonstrates its broad applicability through experiments.
IEEE TRANSACTIONS ON COGNITIVE COMMUNICATIONS AND NETWORKING
(2022)
Article
Computer Science, Information Systems
Ons Aouedi, Kandaraj Piamrat, Benoit Parrein
Summary: This paper proposes a novel approach based on deep learning to tackle the challenge of network traffic classification. By combining decision tree models and deep learning models, the proposed approach aims to improve classification accuracy and has been proven effective in experiments.
IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT
(2022)
Article
Computer Science, Artificial Intelligence
Dunnan Liu, Xiaofeng Xu, Mingguang Liu, Yaling Liu
Summary: With the rapid development of the Internet, network traffic classification has become a hot topic for scientists. The continuous advancement of Internet of Things technology presents new challenges for network traffic classification, where machine learning decision tree classification algorithms have been proven to be efficient and improve network resource utilization.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Ramin Mohammadi, Sedat Akleylek, Ali Ghaffari, Alireza Shirmarz
Summary: This paper discusses the problem of network resource configuration using Software Defined Network (SDN). By building a model that classifies applications based on the type of network flow and optimizes resource allocation, the proposed model improves Quality of Service (QoS) while maximizing network utilization.
CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS
(2023)