pharmine

kNN can be modified for regression problems by replacing the previous equation with the following equation:

The predicted biological value of the given compound is the average of the biological values of its k nearest neighbours. Unlike kNN that is used for classification problems, k need not be an odd number in this case.

This entry was posted by Yap Chun Wei on Saturday, June 14th, 2008 at 10:21 pm and is filed under Data mining, Tutorial. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.