Functions for Naive Bayes

The NAIVEBAYES stored procedure and the PREDICT_NAIVEBAYES stored procedure have the following functions:

• Naive Bayes model building by estimating probabilities
• Support for discrete and continuous attributes
• Automatic discretization of all continuous attributes
• Support column properties definition mechanism
• The following methods of conditional attribute-value probability estimation:
  • Ordinary frequency-based estimation by replacing zero probabilities with small numbers that correspond to half of an instance
  • -estimation with uniform priors and set to the number of attribute values for each attribute
• Prediction of most likely class labels and class probabilities for all classes

All stored procedures consist of a mandatory one-string parameter that contains pairs of <parameter>=<value> entries. These entries are separated by a comma. The data type of the parameter is VARCHAR(any).

Valid <parameter>=<value> entries are listed in the parameter descriptions for each stored procedure.