Teradata analytic functions analyze ordered data. Ordered data is time series data, such as clickstreams, financial transactions, and online user interactions. The following Teradata analytic functions are new for Teradata Database 16.20 Feature Update 1:
| Function Name | Description |
|---|---|
| Attribution | Calculates attributions with a wide range of distribution models. Often used in web-page analysis. |
| DecisionTreePredict | Uses the model generated by the ML Engine DecisionTree function to analyze the input data and make predictions. |
| GLMPredict | Uses the model generated by the ML Engine GLM function to analyze the input data and make predictions. |
| NaiveBayesPredict | Uses the model generated by the ML Engine Naive Bayes Classifier function to analyze the input data and make predictions. |
| NaiveBayesTextClassifierPredict | Uses the model generated by the ML Engine NaiveBayesTextClassifierTrainer function to analyze the input data and make predictions. |
| NPath | Performs regular pattern matching over a sequence of rows from one or more inputs. |
| RandomForestPredict | Uses the model generated by the ML Engine RandomForest function to analyze the input data and make predictions. |
| Sessionize | Maps each click in a clickstream to a unique session identifier. |
| SVMSparsePredict | Uses the model generated by the ML Engine SVMSparse function to analyze the input data and make predictions. |
Additional Information
For more information about Teradata analytic functions, see Teradata Vantage™ - NewSQL Engine Analytic Functions, B035-1206.