Can any one help me out in calculating precision and recall for more than 2 classes. While googling over the internet i found the following help full URL
Calculate Precision and Recall
I want exactly the same available in the above URL using NB classifier except i want precision and recall for more than 2 classes
Please help me
Thank you
There are two different strategies to compute the precision and recall across different classes.
The first option is just to average their values (this is called macro-averaged).
The second option is known as micro-averaging and it is a weighted averaged, where each class importance is defined by its relative number of labelled instances. The more documents in the class the more weight it has.
This article by Yang explains this concepts in detail.
