Hello,

I want to start with some basic data mining algorithms. In MS SQL
2000-AS as a data source a I can use both: relational data and OLAP
data. So I am wondering... having all data in relational database is it
okay to use relational data (directly from that database) to create
data mining models??

Do I lose any of basic functionality, performance, etc. by using
relational data, so that I should rather first create a warehouse from
my database and then create a data mining model based on just created
OLAP data???

Regards, Mark

It's absolutely possible to mine directly from relational data and it's
likely the path many users will take due to the flexibility of that approach
in terms of fewer restrictions, data preparation abilities, use of named
queries and so forth.

Mining on top of OLAP cubes has its own benefits though, including the
ability to take advantage of complex calculated members and the ability to
apply the results of mining back to the source OLAP data in the form of data
mining dimensions.

--
-Raman Iyer
SQL Server Data Mining
[This posting is provided "AS IS" with no warranties, and confers no
rights.]
PLEASE POST NEW QUESTIONS AT
http://forums.microsoft.com/msdn/sho...id=81&siteid=1


<mkazmierski (AT) gmail (DOT) com> wrote