O'Reilly logo

Bioinformatics with R Cookbook by Paurush Praveen Sinha

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

Cross-validation for classifiers

Getting a learning-based model is not enough to get optimal results. The question that remains is how well the model performs when applied to make new predictions for unseen data instances, which we refer to as predictive performance in technical terms. One way is to hold out a part of the available data as a test set. After training, this test set can be used to test the performance of the learned model. This basic idea for a whole class of model evaluation methods is termed cross-validation. Cross-validation (CV) is useful to overcome the problem of overfitting, which refers to a condition where the model requires more information than the data can provide.

There are several approaches to do CV, the simplest being ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required