ETL for Data Mining

Preparing your source data for load into data mining models involves some of the same considerations that you had with data destined for OLAP cubes. Specifically, you’ll want to quality check as best you can. We find two types of business scenarios when loading data mining structures. In some cases, a client prefers to create OLAP cubes first and then use the cleansed data in the star schema or in the cubes themselves as source data for new data mining models. This approach reduces the need for complex ETL specific to the data mining load, because the source data has already been cleansed during the OLAP cube preparation process. However, sometimes we encounter the opposite situation: A client has a huge amount of data that ...

Get Smart Business Intelligence Solutions with Microsoft® SQL Server® 2008 now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.