Naive Bayes Predictor

Predicts the class per row based on the learned model. The class probability is the product of the probability per attribute and the probability of the class attribute itself.

The probability for nominal values is the number of occurrences of the class value with the given value divided by the number of total occurrences of the class value. The probability of numerical values is calculated by assuming a normal distribution per attribute.

Input ports

1. Type: PMML
   The naive Bayes model to use
   A previously learned naive Bayes model
2. Type: Table
   Input data to classify
   Input data to classify

Output ports

1. Type: Table
   The classified data
   The input table with one column added containing the classification and the probabilities depending on the options.