Is Our Ground-Truth for Traffic Classification Reliable?

  • Valentín Carela-Español
  • Tomasz Bujlow
  • Pere Barlet-Ros
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8362)


The validation of the different proposals in the traffic classification literature is a controversial issue. Usually, these works base their results on a ground-truth built from private datasets and labeled by techniques of unknown reliability. This makes the validation and comparison with other solutions an extremely difficult task. This paper aims to be a first step towards addressing the validation and trustworthiness problem of network traffic classifiers. We perform a comparison between 6 well-known DPI-based techniques, which are frequently used in the literature for ground-truth generation. In order to evaluate these tools we have carefully built a labeled dataset of more than 500 000 flows, which contains traffic from popular applications. Our results present PACE, a commercial tool, as the most reliable solution for ground-truth generation. However, among the open-source tools available, NDPI and especially Libprotoident, also achieve very high precision, while other, more frequently used tools (e.g., L7-filter) are not reliable enough and should not be used for ground-truth generation in their current form.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Dainotti, A., et al.: Issues and future directions in traffic classification. IEEE Network 26(1), 35–40 (2012)CrossRefGoogle Scholar
  2. 2.
    Valenti, S., Rossi, D., Dainotti, A., Pescapè, A., Finamore, A., Mellia, M.: Reviewing Traffic Classification. In: Biersack, E., Callegari, C., Matijasevic, M. (eds.) Data Traffic Monitoring and Analysis. LNCS, vol. 7754, pp. 123–147. Springer, Heidelberg (2013)CrossRefGoogle Scholar
  3. 3.
    Fukuda, K.: Difficulties of identifying application type in backbone traffic. In: Int. Conf. on Network and Service Management (CNSM), pp. 358–361. IEEE (2010)Google Scholar
  4. 4.
    Carela-Espñol, V., et al.: Analysis of the impact of sampling on NetFlow traffic classification. Computer Networks 55, 1083–1099 (2011)CrossRefGoogle Scholar
  5. 5.
    Alcock, S., et al.: Libprotoident: Traffic Classification Using Lightweight Packet Inspection. Technical report, University of Waikato (2012)Google Scholar
  6. 6.
    Gringoli, F., et al.: Gt: picking up the truth from the ground for internet traffic. ACM SIGCOMM Computer Communication Review 39(5), 12–18 (2009)CrossRefGoogle Scholar
  7. 7.
    Dainotti, A., et al.: Identification of traffic flows hiding behind TCP port 80. In: IEEE Int. Conf. on Communications (ICC), pp. 1–6 (2010)Google Scholar
  8. 8.
    Karagiannis, T., et al.: Transport layer identification of P2P traffic. In: 4th ACM Internet Measurement Conf. (IMC), pp. 121–134 (2004)Google Scholar
  9. 9.
    Shen, C., et al.: On detection accuracy of L7-filter and OpenDPI. In: 3rd Int. Conf. on Networking and Distributed Computing (ICNDC), pp. 119–123. IEEE (2012)Google Scholar
  10. 10.
    Alcock, S., Nelson, R.: Measuring the Accuracy of Open-Source Payload-Based Traffic Classifiers Using Popular Internet Applications. In: IEEE Workshop on Network Measurements (2013)Google Scholar
  11. 11.
    Dusi, M., et al.: Quantifying the accuracy of the ground truth associated with Internet traffic traces. Computer Networks 55(5), 1158–1167 (2011)CrossRefGoogle Scholar
  12. 12.
    [Online]: Traffic classification at the Universitat Politècnica de Catalunya, UPC (2013),
  13. 13.
    Bujlow, T., et al.: Volunteer-Based System for classification of traffic in computer networks. In: 19th Telecommunications Forum TELFOR, pp. 210–213. IEEE (2011)Google Scholar
  14. 14.
    [Online]: Volunteer-Based System for Research on the Internet (2012),
  15. 15.
    Bujlow, T., et al.: Comparison of Deep Packet Inspection (DPI) Tools for Traffic Classification. Technical report, UPC BarcelonaTech (2013)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Valentín Carela-Español
    • 1
  • Tomasz Bujlow
    • 2
  • Pere Barlet-Ros
    • 1
  1. 1.UPC BarcelonaTechSpain
  2. 2.Aalborg UniversityDenmark

Personalised recommendations