October 2008
Beginner to intermediate
1024 pages
31h 30m
English
SQL Server 2008 improves on its predecessor in a number of areas, including ease of use, feature richness, sophistication of client tools, and programmability. Several routine data mining tasks have now been built into the data mining component of SQL Server 2008, including the creation of holdout test sets, training and testing mining models on subsets of data, and cross validation of models. Whereas in SQL Server 2005 the creation of holdout test sets had to be done outside the data mining tool, SQL Server 2008 incorporates the creation of a test set within the Data Mining Wizard. The holdout test set is then available when testing and comparing the accuracy of all the mining models in the structure. ...
Read now
Unlock full access