"Lars Roith" <lars.roith@nospam.aitag.com> wrote in message
news:%23VirZ3nEEHA.2272@TK2MSFTNGP09.phx.gbl...
> Hello everybody,
>
> I have a relational database containign a table with a lot of facts (e.g.
> yearly income).
> Now I want to build groups/clusters for the yearly income.
> Is there a way to have the analysis services doing the job automtically?
> The result should be:
> e.g 7 Clusters with the following properties:
> income between 0 and 20k
> income between 20k and 30k
> ...
>
> How can I achieve this.
>
> TIA,
>
> Lars
>
>

> This sounds like grouping, not clustering. If on the other hand, you want
> to automatically detect the clusters, you can create a mining model in
> Analysis Services using the clustering algorithm. You would set the
Cluster
> Count to 7 (in the model properties in the Mining Editor) and set the
Income
> property to input. You can either ignore all other columns, or set their
> values to "Predict Only."
>
> Afterwards you can run a query to determine which cluster each customer
was
> assigned - check out the FAQ at
> http://groups.msn.com/AnalysisServicesDataMining for details.
>
> --
>
> -Jamie MacLennan
> SQL Server Data Mining
> This posting is provided "AS IS" with no warranties, and confers no
rights.
> "Lars Roith" <lars.roith@nospam.aitag.com> wrote in message
> news:%23VirZ3nEEHA.2272@TK2MSFTNGP09.phx.gbl...
> > Hello everybody,
> >
> > I have a relational database containign a table with a lot of facts
(e.g.
> > yearly income).
> > Now I want to build groups/clusters for the yearly income.
> > Is there a way to have the analysis services doing the job automtically?
> > The result should be:
> > e.g 7 Clusters with the following properties:
> > income between 0 and 20k
> > income between 20k and 30k
> > ...
> >
> > How can I achieve this.
> >
> > TIA,
> >
> > Lars
> >
> >
>
>