Prediction

Prediction operators follow Modeling operators. Each modeling algorithm operator must be assembled with an input data source for analysis and a Prediction operator for running the model against a new source in order to return a predicted dataset.

There are two general types of Prediction operators.

• Classifier
• Predictor

Classifier Operators

A Classifier calculates a probability of each possible value for the independent variable as well as the value with the highest probability. In addition, it provides confidence values for the model.

Classifiers are terminal operators within database Workflows (no other operator can follow them).

Predictor Operators

A Predictor calculates the following:

• For classification algorithms: the value with the highest probability
• For numeric, regression algorithms: the predicted value

Predictors are not terminal operators within a workflow (other operators can follow them for further analysis of the returned results).

• Prediction and Modeling Operator Pairings
  In general, the Classification Modeling operators are followed in a workflow by a Classifier operator, and the Regression Modeling operators that predict a value are followed by a Predictor Operator.
• Pearson's Chi Square Operations
  Team Studio provides two Chi Square operators for predicting on a Hadoop data set.

Copyright © Cloud Software Group, Inc. All rights reserved.