How to calculate Accuracy, Recall and Precision for multi-class multi-lable Fuzzy inference system in MATLAB?

I've designed a fuzzy inference system in the MATLAB using fuzzy logic toolbox. My target dataset is comprised of 100 instances and this data set is of 21 different classes. Now, I want to calculate its ARP (Accuracy, Recall and Precision) for every class which means there will be 21 different confusion matrix with 21 different ARPs.

I've seen 'plotconfusion' and 'confusionmat' functions of the MATLAB but didn't understand these function. Kindly guide me to create the confusion matrix for my system and how to calculate it in MATLAB.

Answer 1

[Greg Heath]

Use BOTH the help and doc commands on

 confusion
 confusionmat
 plotconfusion
 roc
 plotroc

You can find additional classification examples by using BOTH the help and doc commands on

 nndatasets

If this is not sufficient, search for each of these terms in BOTH NEWSGROUP and ANSWERS

If you still have problems, post your code with accompanying error statements

Hope this helps.

Thank you for formally accepting my answer

Greg