Due to recent updates, all users are required to create an Altair One account to login to the RapidMiner community. Click the Register button to create your account using the same email that you have previously used to login to the RapidMiner community. This will ensure that any previously created content will be synced to your Altair One account. Once you login, you will be asked to provide a username that identifies you to other Community users. Email us at Community with questions.

Unkonw value in distribution table in naive bayes model

cazevedocazevedo Member Posts: 2 Contributor I
edited November 2018 in Help
Hello  :)

I am conducting a study in which the use naive Bayes, and the goal is to get the model, specifically the table of conditional probabilities associated with each variable. However, in the table for each variable appears a class unknow, as the probability of that class. But my base does not contain missing vaues. What can be cause this and how correcting? Because it was not supposed to show that value, only the class values ​​for each variable.

I hope to help me,
Greetings!

Answers

  • MariusHelfMariusHelf RapidMiner Certified Expert, Member Posts: 1,869 Unicorn
    Hey :)

    Try to unset the parameter laplace_corrections in the Naive Bayes operator. Does that help?

    Best regards,
    Marius
  • cazevedocazevedo Member Posts: 2 Contributor I
    Hey Marius  :)

    I did what you said, however the class unknow still appears, but now all these probabilities associated with this class are zero.

    Is there any possibility of this problem have to do with the data import for RapidMiner? I import my data into CSV file, and then in options I pass variables to nominal.

    Thanks for your help!

    Best regards,
    Cátia
  • MariusHelfMariusHelf RapidMiner Certified Expert, Member Posts: 1,869 Unicorn
    Hey Cátia,

    it is normal that the Naive Bayes model always shows this "unknown" category. You can safely ignore it if laplace correction are disabled. It has nothing to do with your input.

    Best regards,
    Marius
Sign In or Register to comment.