Selecting features in microarray classification using ROC curves

by: Hiroshi Mamitsuka

Pattern Recognition, Vol. 39, No. 12. (December 2006), pp. 2393-2404.

View FullText article

DOI, ScienceDirect

Reviews [Write a review of this article]

There are no reviews of this article

Find related articles from these CiteULike users

bigga

Find related articles with these CiteULike tags

feature-selection

Abstract

We present a new method based on the ROC (Receiver Operating Characteristic) curve to efficiently select a feature subset in classifying a high-dimensional microarray dataset with a limited number of observations. Our method has two steps: (1) selecting the most relevant features to the target label using the ROC curve and (2) iteratively eliminating a redundant feature using the ROC curves. The ROC curve is strongly related with a non-parametric hypothesis testing, which must be effective for a dataset with small numerical observations. Experiments with real datasets revealed the significant performance advantage of our method over two competing feature subset selection methods.

@article{citeulike:2799972, abstract = {We present a new method based on the ROC (Receiver Operating Characteristic) curve to efficiently select a feature subset in classifying a high-dimensional microarray dataset with a limited number of observations. Our method has two steps: (1) selecting the most relevant features to the target label using the ROC curve and (2) iteratively eliminating a redundant feature using the ROC curves. The ROC curve is strongly related with a non-parametric hypothesis testing, which must be effective for a dataset with small numerical observations. Experiments with real datasets revealed the significant performance advantage of our method over two competing feature subset selection methods.}, author = {Mamitsuka, Hiroshi }, booktitle = {Bioinformatics}, citeulike-article-id = {2799972}, doi = {10.1016/j.patcog.2006.07.010}, journal = {Pattern Recognition}, keywords = {feature-selection}, month = {December}, number = {12}, pages = {2393--2404}, posted-at = {2008-05-14 21:39:50}, priority = {5}, title = {Selecting features in microarray classification using ROC curves}, url = {http://dx.doi.org/10.1016/j.patcog.2006.07.010}, volume = {39}, year = {2006} }

RIS record

TY - JOUR ID - citeulike:2799972 L3 - citeulike-article-id:2799972 TI - Selecting features in microarray classification using ROC curves T2 - Bioinformatics JF - Pattern Recognition VL - 39 IS - 12 SP - 2393 EP - 2404 N2 - We present a new method based on the ROC (Receiver Operating Characteristic) curve to efficiently select a feature subset in classifying a high-dimensional microarray dataset with a limited number of observations. Our method has two steps: (1) selecting the most relevant features to the target label using the ROC curve and (2) iteratively eliminating a redundant feature using the ROC curves. The ROC curve is strongly related with a non-parametric hypothesis testing, which must be effective for a dataset with small numerical observations. Experiments with real datasets revealed the significant performance advantage of our method over two competing feature subset selection methods. KW - feature-selection AU - Mamitsuka, Hiroshi PY - 2006/12// UR - http://dx.doi.org/10.1016/j.patcog.2006.07.010 ER -

RIS BibTeX