Unsupervised Non-redundant Feature Selection: A Graph-Theoretic Approach

Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 199)

Abstract

In this article a graph-theoretic approach for non-redundant unsupervised feature selection has been presented. The input data matrix is first converted into a weighted undirected complete feature-graph where the nodes represent the features and the edges are weighted according to the dissimilarity of features. Then the densest subgraph having maximum average weight is identified from the original feature graph. The features contained in the reduced subgraph are the final selected features for which average correlation is very less. The proposed method is compared with other dimensionality reduction techniques such as SFS and SBS in terms of entropy, classification accuracy, class separability, average correlation and execution time on several real life data sets.

Keywords

Filter Unsupervised Entropy Class Separability 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Kohavi, R., John, G.: Wrapper for feature subset selection. Artificial Intelligence 97, 273–324 (1997)CrossRefMATHGoogle Scholar
  2. 2.
    Jiang, S., Wang, L.: An unsupervised feature selection framework based on clustering. School of Informatics, Guangdong University of Foreign Studies, Guangzhou (2008)Google Scholar
  3. 3.
    Cai, D., Zhang, C., He, X.: Unsupervised feature selection for multi-cluster data. In: KDD 2010, Washington, DC, USA (2010)Google Scholar
  4. 4.
    Dy, J.G., Brodley, C.E., Kak, A., Broderick, L.S., Aisen, A.M.: Unsupervised feature selection applied to content-based retrieval of lung images. IEEE Transaction on Pattern Analysis and Machine Intellegence 25(3), 373–378 (2003)CrossRefGoogle Scholar
  5. 5.
    Morita, M., Oliveira, L.S., Sabourin, R.: Unsupervised feature selection for ensemble of classifiers. Frontiers in Handwriting Recognition (2004)Google Scholar
  6. 6.
    Zhang, Z., Hancock, E.R.: A graph-based approach to feature selection. Springer (2011)Google Scholar
  7. 7.
    Bahmani, B., Kumar, R., Vassilvitskii, S.: Densest subgraph in streaming and mapreduce. VLDB Endowment 5(5), 454–465 (2012)Google Scholar
  8. 8.
    Mitra, P., Murthy, C.A., Pal, S.K.: Unsupervised feature selection using feature similarity. IEEE Transaction on Pattern Analysis and Machine Intellegence 24(3), 301–312 (2002)CrossRefGoogle Scholar
  9. 9.
    Li, Y., Lu, B., Wu, Z.: A hybrid method of unsupervised feature selection based on ranking. IEEE Computer Society, Washington, DC (2006)Google Scholar
  10. 10.
    Sondberg-Madsen, N., Thomsen, C., Pena, J.M.: Unsupervised feature subset selection. In: Proceedings of the Workshop on Probabilistic Graphical Models for Classification (2003)Google Scholar
  11. 11.
    Chatterjee, S., Hadi, A.S.: Regression Analysis by Example(4e). John Wiley & Sons, Inc.Google Scholar
  12. 12.
    Dash, M., Liu, H.: Unsupervised feature selection. In: Proc. Pacific Asia Conf. Knowledge Discovery and Data Mining (2000)Google Scholar
  13. 13.
    Devijver, P.A., Kittler, J.: Pattern recognition: A statistical approach. Prentice-Hall, Englewood Cliffs (1982)MATHGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  1. 1.Department of Computer Science and EngineeringUniversity of KalyaniKalyaniIndia

Personalised recommendations