Optimality and stability of feature set for traffic classification

Optimality and stability of feature set for traffic classification

For access to this article, please select a purchase option:

Buy chapter PDF
(plus tax if applicable)
Buy Knowledge Pack
10 chapters for $120.00
(plus taxes if applicable)

IET members benefit from discounts to all IET publications and free access to E&T Magazine. If you are an IET member, log in to your account and the discounts will automatically be applied.

Learn more about IET membership 

Recommend Title Publication to library

You must fill out fields marked with: *

Librarian details
Your details
Why are you recommending this title?
Select reason:
Network Classification for Traffic Management: Anomaly detection, feature selection, clustering and classification — Recommend this title to your library

Thank you

Your recommendation has been sent to your librarian.

Feature selection (FS) methods can be used as a preprocessing step to eliminate meaningless features, and also as a tool to reveal the set of optimal features. Unfortunately, as detailed in Chapter 6, such methods are often sensitive to a small variation in the traffic data collected over different periods of time. Thus, obtaining a stable feature set is crucial in enhancing the confidence of network operators. This chapter describes a robust approach, called global optimization approach (GOA), to identify both optimal and stable features, relying on a multi-criterion fusion-based FS method and an information-theoretic method. GOA first combines multiple well-known FS methods to yield possible optimal feature subsets across different traffic datasets and then uses the proposed adaptive threshold, which is based on entropy to extract the stable features. A new goodness measure is proposed within a random forest framework to estimate the final optimum feature subset. The effectiveness of GOA is demonstrated through several experiments on network traffic data in spatial and temporal domains. Experimental results show that GOA provides up to 98.5% accuracy, exhibits up to 50% reduction in the feature set size, and finally speeds up the runtime of a classifier by 50% compared with individual results produced by other well-known FS methods.

Chapter Contents:

  • 7.1 Introduction
  • 7.2 Optimality versus stability
  • 7.2.1 Selecting feature set from global perspective
  • 7.2.2 An initial investigation
  • 7.3 GOA—the global optimization approach
  • 7.3.1 Integration of feature selection
  • 7.3.2 The adaptive threshold
  • 7.3.3 Intensive search approach
  • 7.4 Evaluation
  • 7.4.1 GOA evaluation based on the proposed metrics
  • 7.4.2 Comparison between GOA, FCBF-NB, and BNN
  • 7.4.3 Relevance of selected features
  • 7.4.4 Temporal decay and spatial robustness
  • 7.5 Impact of the candidate features on different ML algorithms
  • 7.5.1 The sensitivity of the candidate features on different ML algorithms
  • 7.5.2 Discretization to improve classification accuracy
  • 7.5.3 Impact of discretizing the candidate features
  • 7.6 Conclusion

Inspec keywords: feature extraction; entropy; optimisation; telecommunication traffic; data analysis; feature selection; pattern classification

Other keywords: feature selection methods; FS methods; meaningless features; preprocessing step; traffic classification; feature set size; GOA; optimal features; global optimization approach; stable feature set; optimal feature subsets; network traffic data; multicriterion fusion-based FS method; final optimum feature subset; traffic datasets; information-theoretic method

Subjects: Data handling techniques; Optimisation techniques

Preview this chapter:
Zoom in

Optimality and stability of feature set for traffic classification, Page 1 of 2

| /docserver/preview/fulltext/books/pc/pbpc032e/PBPC032E_ch7-1.gif /docserver/preview/fulltext/books/pc/pbpc032e/PBPC032E_ch7-2.gif

Related content

This is a required field
Please enter a valid email address