Analysis and Classification of Encrypted Network Traffic Using Machine Learning

Muliukha, Vladimir; Laboshin, Leonid; Lukashin, Alexey; Nashivochnikov, Nikolay V.

doi:10.1109/scm50615.2020.9198811

Cited by 13 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, with our 12-feature decision tree model, we obtained an overall accuracy of 99.76%, which represents fewer features for more performance. We also obtained higher overall accuracies from the other models despite requiring more features (99.75% for the 17 [19]. Results and ours when using the eleventh block.…”

Section: Comparison Of Our Results With the State-of-the-artsupporting

confidence: 56%

“…One particular study reviewed current traffic classification methods by classifying them into five categories: statistics-based, correlation-based, behaviour-based, payload-based, [15]. Some studies [16], [17] have provided classification methods for encrypted traffic, which was challenging to perform in the past. Today port-based analysis is ineffective, being unable to identify 30-70% of today's internet traffic [5], [11].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Features analysis of internet traffic classification using interpretable machine learning models

Adje

Houndji²,

Dossou³

2022

IJ-AI

View full text Add to dashboard Cite

Internet traffic classification is a fundamental task for network services and management. There are good machine learning models to identify the class of traffic. However, finding the most discriminating features to have efficient models remains essential. In this paper, we use interpretable machine learning algorithms such as decision tree, random forest and eXtreme gradient boosting (XGBoost) to find the most discriminating features for internet traffic classification. The dataset used contains 377,526 traffics. Each traffic is described by 248 features. From these features, we propose a 12-feature model with an accuracy of up to 99.76%. We tested it on another dataset with 19626 flows and obtained 98.40% of accuracy. This shows the efficiency and stability of our model. Also, we identify a set of 14 important features for internet traffic classification, including two that are crucial: port number (server) and minimum segment size (client to server).

show abstract

Section: Comparison Of Our Results With the State-of-the-artsupporting

confidence: 56%

Section: Introductionmentioning

confidence: 99%

Features analysis of internet traffic classification using interpretable machine learning models

Adje

Houndji²,

Dossou³

2022

IJ-AI

View full text Add to dashboard Cite

show abstract

“…Additionally, ensemble methods such as Random Forest and Gradient Boosted Trees have received attention for their robustness and accuracy in handling complex datasets with multiple input variables. These methods combine multiple decision trees to improve the predictive performance and have proven to be particularly effective in enhancing the detection capabilities within VPN security frameworks (Wang, Z. et al 2022), (Muliukha, V.A. et al 2020).…”

Section: Literature Reviewmentioning

confidence: 99%

“…Among the models tested, ensemble methods, particularly Random Forest and Gradient Boosting Trees, were highlighted for their superior performance in terms of accuracy and the ability to handle overfitting (Bagui, S. et al 2017). Muliukha et al (2020) discussed the application of machine learning algorithms such as Random Forest and Naive Bayesian for the classification of encrypted network traffic, specifically focusing on SSL sessions and VPN connections. Similarly, Wang et al (2022) provided a comprehensive analysis of using machine learning to detect encrypted malicious traffic, offering insights into different datasets and the efficacy of various ML algorithms in this context.…”

Section: Literature Reviewmentioning

confidence: 99%

AI-Enhanced VPN Security Framework: Integrating Open-Source Threat Intelligence and Machine Learning to Secure Digital Networks

Hasan,

Malik

2024

eccws

View full text Add to dashboard Cite

In today's digital age, ensuring network privacy and integrity is of utmost importance. To address this, our work proposed an advanced VPN security framework that integrates open-source threat intelligence and machine learning (ML) to enhance cyber defences. By combining Wazuh for threat detection and analysis, and pfsense for firewall capabilities, with state-of-the-art ML algorithms, we present a robust VPN security solution to the challenges presented by the evolving landscape of cyber threats, representing a significant advancement in securing digital networks. This framework is strengthened by the integration of four ML algorithms— Gradient Boosted Trees (GBT), Random Forest (RF), K-Nearest Neighbors (KNN), and Dense Deep Learning (DDL)— chosen for their classification efficacy and their ability to process complex security data, thereby improving the efficiency and accuracy of threat detection. Results indicated significant improvements in threat detection accuracy following the integration of ML algorithms. The Random Forest (RF) algorithm, in particular, stood out for its exceptional accuracy and ability to handle various threat scenarios, showcasing its efficacy in identifying sophisticated cyber threats through network traffic pattern analysis. Further performance benchmarking confirmed the feasibility of deploying the advanced VPN security framework, demonstrating minimal impact on network latency and throughput.

show abstract

Analysis of Classification Methods for Wi-Fi Signals Based on the Use of Channel State Information Spatial Features and CNN

Lopatin

Fyodorov

2022

Springer Proceedings in Physics

View full text Add to dashboard Cite

Analysis and Classification of Encrypted Network Traffic Using Machine Learning

Cited by 13 publications

References 9 publications

Features analysis of internet traffic classification using interpretable machine learning models

Features analysis of internet traffic classification using interpretable machine learning models

AI-Enhanced VPN Security Framework: Integrating Open-Source Threat Intelligence and Machine Learning to Secure Digital Networks

Analysis of Classification Methods for Wi-Fi Signals Based on the Use of Channel State Information Spatial Features and CNN

Contact Info

Product

Resources

About