Groups | Blog | Home
all groups > sql server data mining > march 2004 >

sql server data mining : Angoss SDK sample - Interpreting the clusters?


Peter Kim [MS]
3/23/2004 10:21:47 AM
My experience in the Angoss control is limited, but a good starting point of
the investigation may be getting the cluster content schema rowset and
compare with the chart. The cluster content schema rowset can be obtained by
DMSample using the following DMX query:

SELECT * FROM <your_clustering_model>.CONTENT

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

[quoted text, click to view]

Sinex
3/23/2004 5:25:06 PM
Hi !
I tried the Angoss sample application that comes with the SDK. The
cluster model shows up in the SegView control.
The first column makes sense - it talks abt the distinct clusters in the
entire population. But wat are the other pie charts? How was cluster 001 to
cluster 010 created? Definitely it must be based on some other field (say
the IQ or ParentIncome)...but that is not coming out well in the pie chart
:-( And wat is that probability distribution sort of curve ? HELP !

Sinex

AddThis Social Bookmark Button