Advertisement

Improving the Dendritic Cell Algorithm Performance Using Fuzzy-Rough Set Theory as a Pattern Discovery Technique

  • Zeineb Chelly
  • Zied Elouedi
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 303)

Abstract

As an immune inspired algorithm, the Dendritic Cell Algorithm (DCA) is based on the behavior of biological dendritic cells. The performance of DCA relies on its data pre-processing phase; including feature selection and signal categorization. For an automatic data pre-processing task, DCA applied Rough Set Theory (RST). However, applying RST as a pre-processor presents an information loss as data should be discretized beforehand. Therefore, the aim of this paper is to propose a new DCA data pre-processing phase based on a more efficient pattern discovery technique which is Fuzzy Rough Set Theory (FRST). FRST allows dealing with real-valued data with no data quantization beforehand. In our newly proposed fuzzy-rough model, the data pre-processing phase is based on FRST and more precisely on the use of the fuzzy positive region and the fuzzy dependency degree concepts. The experimental results show that our proposed algorithm succeeds in obtaining significantly improved classification accuracy.

Keywords

Artificial Immune Systems Dendritic Cells Fuzzy-Rough Set Theory Pattern Discovery 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Eiben, A., Smith, J.: Introduction to Evolutionary Computing, 2nd edn. Natural Computing (2007)Google Scholar
  2. 2.
    Greensmith, J., Aickelin, U., Cayzer, S.: Introducing dendritic cells as a novel immune-inspired algorithm for anomaly detection. In: Jacob, C., Pilat, M.L., Bentley, P.J., Timmis, J.I. (eds.) ICARIS 2005. LNCS, vol. 3627, pp. 153–167. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  3. 3.
    Chelly, Z., Elouedi, Z.: Qr-dca: A new rough data pre-processing approach for the dendritic cell algorithm. In: Tomassini, M., Antonioni, A., Daolio, F., Buesser, P. (eds.) ICANNGA 2013. LNCS, vol. 7824, pp. 140–150. Springer, Heidelberg (2013)CrossRefGoogle Scholar
  4. 4.
    Pawlak, Z.: Rough sets. International Journal of Computer and Information Science 11, 341–356 (1982)CrossRefMATHMathSciNetGoogle Scholar
  5. 5.
    Jensen, R., Shen, Q.: New approaches to fuzzy-rough feature selection. IEEE Transactions on Fuzzy Systems 17, 824–838 (2009)CrossRefGoogle Scholar
  6. 6.
    Asuncion, A., Newman, D.J.: UCI machine learning repository (2007), http://mlearn.ics.uci.edu/mlrepository.html
  7. 7.
    Chelly, Z., Elouedi, Z.: Rst-dca: A dendritic cell algorithm based on rough set theory. In: Huang, T., Zeng, Z., Li, C., Leung, C.S. (eds.) ICONIP 2012, Part III. LNCS, vol. 7665, pp. 480–487. Springer, Heidelberg (2012)CrossRefGoogle Scholar
  8. 8.
    Chelly, Z., Elouedi, Z.: Rc-dca: A new feature selection and signal categorization technique for the dendritic cell algorithm based on rough set theory. In: Coello Coello, C.A., Greensmith, J., Krasnogor, N., Liò, P., Nicosia, G., Pavone, M. (eds.) ICARIS 2012. LNCS, vol. 7597, pp. 152–165. Springer, Heidelberg (2012)CrossRefGoogle Scholar
  9. 9.
    Chelly, Z., Elouedi, Z.: A new data pre-processing approach for the dendritic cell algorithm based on fuzzy rough set theory. In: Proceedings of the Genetic and Evolutionary Computation Conference, GECCO 2013, pp. 163–164. ACM (2013)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  1. 1.High Institute of Management of TunisLARODEC, University of TunisTunisTunisia

Personalised recommendations