Typical usages of clustering are:

- Identify distinct clusters in the data (e.g., customers):
- the cluster id for each case can be obtained by Cluster() - see FAQ,
item 14 for an example.
- cluster viewer can give profile information on each cluster, which
will help understand a particular cluster. The profile information is also
available from MINING_MODEL_CONTENT schema rowset as well.

- Further data mining for each cluster
- Using DTS prediction task, write out the data with cluster id - again,
see FAQ item 14 for the example.
- For each cluster, build a decision tree to predict attributes
interested (e.g., Card Type). Each cluster might have different pattern to
predict the attributes.
- Use the decision tree model to score existing customers in batch
(using PREDICTION JOIN with OPENROWSET) or real time (using PREDICTION JOIN
with singleton SELECT).

--
Peter Kim
This posting is provided "AS IS" with no warranties, and confers no rights.

[quoted text, click to view]

Hi !
I have a cluster model of my data. I'd like to perform queries against
this model now. What sort of queries can I make? How can i make more sense
of the clusters?

....Sinex