Feature selection using Haar wavelet power spectrum
2006

Feature Selection Using Haar Wavelet Power Spectrum

Sample size: 72 publication Evidence: moderate

Author Information

Author(s): Subramani Prabakaran, Sahu Rajendra, Verma Shekhar

Primary Institution: ABV-Indian Institute of Information Technology and Management, Gwalior, India

Hypothesis

The aim of this paper is to assess the capability of Haar wavelet power spectrum in the problem of clustering and gene selection based on expression data in the context of disease classification.

Conclusion

The proposed method for feature selection using Haar wavelet power spectrum is simple, faster, and effective for a wide range of datasets.

Supporting Evidence

  • The Haar wavelet power spectrum method selected most of the genes identified by previous complex methods.
  • The technique was validated against noise in the data to test its robustness.
  • The method is simpler and requires no special software for implementation.
  • Top genes selected were found to be dominant in their respective diagnostic categories.

Takeaway

This study shows a new way to pick important genes from a lot of data using a special math tool called Haar wavelet, which helps in understanding diseases better.

Methodology

The study used Haar wavelet transforms to analyze gene expression data and select informative features for classification.

Limitations

The method was tested without filtering noise or outliers, which may affect results.

Digital Object Identifier (DOI)

10.1186/1471-2105-7-432

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication