1.0 - 8.00 - ConfusionMatrix - Teradata Vantage

Teradata® Vantage Machine Learning Engine Analytic Function Reference

Teradata Vantage
Release Number
Release Date
May 2019
Content Type
Programming Reference
Publication ID
English (United States)

The ConfusionMatrix function shows how often a classification algorithm correctly classifies items. The function takes an input table that includes two columns—one containing the observed class of an item and the other containing the class predicted by the algorithm—and outputs three tables:

  • A confusion matrix (also called a contingency table), which shows the performance of the algorithm
  • A table of overall statistics
  • A table of statistics for each class