need to calculates the Posterior probability using fitcnb function classification

I need quick help with modifications of the attached code which calculates the prior probability and I need to add a function that calculates the Posterior probability of the data using fitcnb classification function

I tried with to apply similar code of this example but I couldn't fit it into my code

https://www.mathworks.com/help/stats/compactclassificationnaivebayes.predict.html#buf5w06-2

these are the classes inside my dataset

'BENIGN','DrDoS_DNS','DrDoS_LDAP','DrDoS_MSSQL','DrDoS_NetBIOS','DrDoS_NTP','DrDoS_SNMP','DrDoS_SSDP','DrDoS_UDP','Syn','UDP-lag'

where the distribution of BENIGN label should be 50% and the other 50% should be all other labels

Thanks in advance realy appreciate your help

Answer 1

[Avadhoot]

Hi Ibtihal,

I understand that you have already created a Naïve Bayes Classifier using the “fitcnb” function. To calculate the posterior probabilities, you need to add the following line to your code after the creation of the model.

[labels,PostProbs,MisClassCost] = predict(Mdl,xdata);

The variables in the above line of code contain the following information:

• “Mdl”: Naïve Bayes classifier model.
• “xdata”: Input data for which you want to calculate the posterior probabilities.
• “PostProbs”: Posterior probabilities of all the classes in your case.
• “labels”: The predicted labels for your input data.
• “MisClassCost”:Misclassification costs for all the classes in each data point.

For more information on how the posterior probabilities and misclassification costs are estimated, refer to the following link:

https://www.mathworks.com/help/releases/R2022a/stats/compactclassificationnaivebayes.predict.html#budvxv0-11:~:text=confusionchart(YTest%2Clabel)%3B-,Estimate,-Posterior%20Probabilities%20and

Regards,

Avadhoot.