Traffic classification using clustering algorithms

By Jeffrey Erman, Martin F. Arlitt, Anirban Mahanti. Annual ACM Workshop on Mining Network Data, 2006

790 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Networks and Communications

View Paper: https://www.semanticscholar.org/paper/9ce3e9ca5c0970ddb32004028d706574d6f64781

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Networks and Communications

Citations in the corpus

Internet traffic classification demystified: myths, caveats, and the best practices (2008, methodology): “Erman et al. [ 16 ] compared the performance of unsupervised machine learning algorithms in traffic classification.”
A survey of techniques for internet traffic classification using machine learning (2008, methodology): “Three accuracy metrics were used for evaluation: recall, precision and overall accuracy (overall accuracy is defined the same as [45] reviewed in the previous sections).”
Network Traffic Classification Using Correlation Information (2013, methodology): “[27] evaluated the k-means, DBSCAN and AutoClass algorithms for traffic clustering on two empirical data traces.”
Clustering algorithms: A comparative approach (2016, methodology): “This was observed for all algorithms and is in accordance with previous results [44].”
State-of-the-Art Deep Learning: Evolving Machine Intelligence Toward Tomorrow’s Intelligent Network Traffic Control Systems (2017, methodology): “In order to remedy the issue, in [176], Erman et al.”
A comprehensive survey on machine learning for networking: evolution, applications and research opportunities (2018, background): “[135] leverage DBSCAN with transport layer statistics to identify application types.”
Unsupervised Machine Learning for Networking: Techniques, Applications and Research Challenges (2017, methodology): “[171] K-Means, DBSCAN Data clustering Applied cluster analysis to effectively identify similar traffic using transport layer statistics to overcome the problem of dynamic port allocation in port based classification.”
Offline/realtime traffic classification using semi-supervised learning (2007, methodology): “The hypothesis here is that the clustering step produces “pure” (in the sense of application types) clusters; in our earlier work [10], we provided empirical evidence of this hypothesis.”