Marcel Brun1, Qian Xu, Edward R Dougherty
1Computational Biology Division, Translational Genomics Research Institute, Phoenix, AZ 85004, USA.
You might also read
Articles linked to this work by shared authors, journal, and citation graph.
Designing a classifier using the full sample dataset consistently yields better results than using a holdout subset for error estimation. Full-sample design offers a superior classifier with a smaller expected error bound.
Area of Science:
Background:
Purpose of the Study:
Main Methods:
Main Results:
Conclusions: