I'm building a classifier of multi-class problem. In my context I have to classify vehicles into five categories: cars, vans, trucks, buses and motorcycles. I would like to use some measure to represent the classification performance for each class and for all of them. In the samples I have the motorcycles are few meanwhile there are a lot of cars so the classes are clearly unbalanced. I read that the Matthews correlation coefficient (MCC) is better in this case but I didn't find any reference that confirm this.

Please, could you help me on this I'm a little bit lost and I don't know which measure to use.


https://eva.fing.edu.uy/pluginfile.php/69453/mod_resource/content/1/7633-10048-1-PB.pdf I think this will be a good reference for you.

To my understanding, F-measure is a "mean" of precision and recall (2*precision*recall/(precision+recall)). For an unbalanced testing set, precision, recall, or F-measure could be confusing and misleading because a random guessing would make F-measure higher than 0.5. But MCC will give you a value around 0 in this situation. So I think MCC is a better choise in your case.

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.