Feature Selection

  • Nishchal K. VermaEmail author
  • Al Salour
Part of the Studies in Systems, Decision and Control book series (SSDC, volume 256)


As detailed in Chap.  4, features have been extracted from the pre-processed data. Too many features may lead to the curse of dimensionality issues. This chapter explains how to obtain a set of relevant features which is the process of feature selection. Feature selection is the process in which most informative variables are selected for the generation of the model. It helps to remove the redundant data and contributes to proper classification. While minimizing the redundancy, we should keep in mind that the predicted information must be preserved as much as possible. This chapter describes various feature selection methods such as Principal Component Analysis (PCA)-based approach, Mutual Information (MI), Bhattacharyya Distance (BD), and Independent Component Analysis (ICA). A novel feature selection based on graphical review has also been elaborated later in this chapter.


  1. 1.
    Verma, N.K., Maini, T., Salour, A.: Acoustic signature based intelligent health monitoring of air compressors with selected features. In: 9th International Conference on Information Technology: New Generations, Nevada, USA, pp. 839–845 (2012)Google Scholar
  2. 2.
    Shlens, J.: A Tutorial on Principal Component Analysis. University of California, Systems Neurobiology Lab., California, USA (2005)Google Scholar
  3. 3.
    Thirukovalluru, R., Sevakula, R.K., Dixit, S., Verma, N.K.: Generating optimum feature sets for fault diagnosis using denoising stacked auto-encode. In: IEEE International Conference on Prognostics and Health Management, Canada, USA, pp. 1–7 (2016)Google Scholar
  4. 4.
    Verma, N.K, Gupta, V.K., Sharma, M., Sevakula, R.K.: Intelligent condition based monitoring of rotating machines using sparse auto-encoders. In: IEEE Conference on Prognostics and Health Management, Maryland, USA, pp. 1–7 (2013)Google Scholar
  5. 5.
    Maurya, S., Singh, V., Dixit, S., Verma, N.K., Salour, A., Liu, J.: Fusion of low-level features with stacked autoencoder for condition based monitoring of machines. In: IEEE International Conference on Prognostics and Health Management, Washington, USA, pp. 11–13 (2018)Google Scholar
  6. 6.
    Sharma, A.K., Singh, V., Verma, N.K., Liu, J.: Condition based monitoring of machine using Mamdani fuzzy network. In: Prognostics and System Health Management Conference, Chongqing, China, pp. 26–28 (2018)Google Scholar
  7. 7.
    Verma, N.K., Dixit, S., Sevakula, R.K., Salour, A.: Computational framework for machine fault diagnosis with autoencoder variants. In: 2018 International Conference on Sensing, Diagnostics, Prognostics, and Control, Xi’an, China, pp. 15–17 (2018)Google Scholar
  8. 8.
    Jolliffe, I.T.: Principal Component Analysis. Springer, New York (1986)CrossRefGoogle Scholar
  9. 9.
    He, Q., Yan, R., Kong, F., Du, R.: Machine condition monitoring using principal component representations. J. Mech. Syst. Signal Pr. 23(2), 446–466 (2009)CrossRefGoogle Scholar
  10. 10.
    Battiti, R.: Using mutual information for selecting features in supervised neural net learning. IEEE Trans. Neural Netw. 5(4), 537–550 (1994)CrossRefGoogle Scholar
  11. 11.
    Peng, H., Long, F., Ding, C.: Feature selection based on mutual information: criteria of max dependency, max-relevance and min-redundancy. IEEE Trans. Pattern Anal. Mach. Intell. 27(8), 1226–1238 (2005)Google Scholar
  12. 12.
    Kwak, N., Choi, C.H.: Input feature selection for classification problems. IEEE Trans. Neural Netw. 3(1), 143–159 (2002)CrossRefGoogle Scholar
  13. 13.
    Estevez, P.A., Tesmer, M., Perez, C.A., Zurada, J.M.: Normalized mutual information feature selection. IEEE Trans. Neural Netw. 20(2), 189–201 (2009)CrossRefGoogle Scholar
  14. 14.
    Guorong, X., Peiqi, C., Minhui, W.: Bhattacharyya distance feature selection. In: Proceedings 13th IEEE International Conference on Pattern Recognition, 2, pp. 195–199 (1996)Google Scholar
  15. 15.
    Aapo, H., Karhunen, J., Oja, E.: Independent Component Analysis, vol. 46. Wiley (2004)Google Scholar
  16. 16.
    Verma, N.K., Sevakula, R.K., Thirukovalluru, R.: Pattern analysis framework with graphical indices for condition based monitoring. IEEE Trans. Rel. 66(4), 1085–1100 (2017)CrossRefGoogle Scholar
  17. 17.
    Verma, N.K., Sevakula, R.K., Dixit, S., Salour, A.: Intelligent condition based monitoring using acoustic signals for air compressors. IEEE Trans. Rel. 65(1), 291–309 (2016)CrossRefGoogle Scholar
  18. 18.
    Verma, N.K., Singh, S., Gupta, J.K., Sevakula, R.K., Dixit, S., Salour, A.: Smartphone application for fault recognition. In: 6th International Conference on Sensing Technology, Kolkata, India, pp. 1–6 (2012)Google Scholar
  19. 19.
    Verma, N.K., Singh, J.V., Gupta, M., Sevakula, R.K., Dixit, S.: Windows mobile and tablet app for acoustic signature machine health monitoring. In: International Conference on Industrial and Information Systems, Gwalior, India, pp. 1–6 (2014)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  1. 1.Department of Electrical Engineering and Inter-disciplinary Program in Cognitive ScienceIndian Institute of Technology KanpurKanpurIndia
  2. 2.Boeing Research and TechnologySaint LouisUSA

Personalised recommendations