Pages: [1]
Author Topic: Prior probability and Naive Bayes  (Read 3667 times)
Posts: 1

« on: June 18, 2013, 03:54:49 AM »

Hi all,

Thank you for taking the time to read this question! Is it possible to create a Naive Bayes classifier from the prior probabilities? I do not have a training set per se, and I was hoping to not include the correct outcome in the data. So, I was hoping to predict the outcome from the prior probabilities, not from a training set. Can this be done in RapidMiner?

On a related note, is it possible to include the prior probabilities of the outcomes? If, for example, I know that 80% of the outcomes will be of one type and the remaining 20% will be the other, is it possible to enter this into the program?

Thank you,

Marius Helf
Hero Member
Posts: 1811

« Reply #1 on: June 20, 2013, 12:23:49 PM »

What you want is probably not possible. RapidMiner usually needs training data to create any kind of model. Maybe you can create a rule-based process with the Branch operator or the Generate Attributes operator, but that will be a bit tedious.

Best regards,

Please add [SOLVED] to the topic title when your problem has been solved! (do so by editing the first post in the thread and modifying the title)
Please click here before posting.
Pages: [1]
Jump to: