5.4.2 - Overview of Variable Transformation - Teradata Warehouse Miner

In-Database Analytic Functions User Guide

Teradata Warehouse Miner
October 2016
User Guide

The Variable Transformation analysis reads a single table or view as input and produces a SELECT statement, Create Table or Create View statement containing transformed columns. This is particularly useful when preparing data for input to an analytic algorithm. For example, a K-Means clustering algorithm will typically produce better results when the input columns are first converted to their Z-Score values to put all input variables on an equal footing, regardless of their magnitude. In classic Teradata Warehouse Miner you use a series of dialogs to collect parameter information, the SQL for the requested transformations is generated, and the SQL is executed. In the in-database version of Variable Transformation, which is independent of other applications, you define the desired transformations in an external stored procedure call statement. Transformations can be performed without the need of the Teradata Warehouse Miner client application, provided that the external stored procedure (td_analyze) is installed and the necessary permissions granted.