What is the prior probability P(C = ‘no’) ? how i can find prior probability in this table ? Naïve


  • Options
    BalazsBaranyBalazsBarany Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert Posts: 955 Unicorn

    in the RapidMiner Academy there are multiple videos about Naive Bayes.

    Here's RapidMiner founder Ingo explaining the algorithm:

    If you look at the model description, it says:


    Distribution model for label attribute Play
    Class no (0.357143)
    4 distributions
    Class yes (0.642857)
    4 distributions
    These are exactly the fractions of the values in the entire distribution - the prior probability of the class in the training data.

    The numbers you see in the table can be interpreted as "the fraction of the data where Outlook = overcast and Play = yes". 

  • Options
    Telcontar120Telcontar120 Moderator, RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,635 Unicorn
    If you look at the table, you can see that the probabilities shown will total to 1 (not accounting for rounding) for each class for each attribute.  So there are 4 possible values for the outlook attribute (rain, overcast, sunny, unknown) and 2 classes (yes/no) and the sums for each set will total to 1. 
    Brian T.
    Lindon Ventures 
    Data Science Consulting from Certified RapidMiner Experts
Sign In or Register to comment.