Chapter 3Prediction Queries with Decision Trees

This chapter demonstrates how to perform DMX Prediction queries with mining models based on the Decision Trees algorithm. Cases queries reveal the original source data for the model. Content queries show the result of training the model and the patterns and clusters and correlations discovered. Both Cases and Content queries are based on existing data. Prediction queries, on the other hand, work on new data (one exception to this rule is the Time Series algorithm, which usually makes predictions based on existing data). They do so by comparing the new data with the results discovered during model training. For example, a Prediction query might show the possibility and probability of a new customer’s ...

Get Practical DMX Queries for Microsoft SQL Server Analysis Services 2008 now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.