Thanks in advance for the help.
MCC gives a measure of the quality of a binary classifier. I'm looking for a similar measure that can be used for a multi-class classifier. Ultimately what I would like to do is to compare several multi-class classifiers that I have created. I could simply look at percent correct, but I would like a secondary measure since percent correct can sometimes be misleading (especially in cases of class imbalance).
Best Answer
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0041882
They have the definition of MCC for the multiclass case. It is easy to implement it by yourself based on eq.2