I would like to have some feedback on a few questions regarding finding anomalities and strange transactions in a dataset.
I have a very simple dataset consisting of three discrete attributes and one continuous value, and the transaction date like this:

What is the easiest way of getting the result of a prediction query saved into a table using only SQL? I want to build a normal database stored procedure, that executes a DMX statement, and saves the result into a table.
/Erik

I would like to know what is the meaning (and how is computing) the field "Favors" into the Clustering Viewer (last panel).
I don't understand the meaning of these values.
Thank you in advance
Victor

In practice is there any guidelines about using data normalization.  Non of the MS documentation mentions anything about this topic, but in some literature, it says it is commonly effective in helping the model.  Is normallization commonly used in industry?

I'm still new to this DM thing (I'm running on SQL2000). I've tried to follow the tutorial, but odd things happen. E.g. when I try to save the cube after some changes, it comes up with an error stating that the object / database is read-only. Also, after exiting AS and then ...

Where can I find more details about Microsoft Data mining Algorithms? For example I would like to know how the Time series algorithm works...
Thank you

I've implemented a time series algorithm over an OLAP cube using the "SQL Server BI Dev Studio".
Using the "Data Mining Wizard" I've managed only to make a mining structure WITH nested-tables, although I've seen DMX examples of creating a mining-structure for time-series WITHOUT ...

I have been requested to represent the results of a Neural Nets and Naive Bayes models in the R2, ROC format.
Is there any way to derive a number that would closely represent a R2 regression coeficient?
Thanks,
Peter

I installed OLE DB for DM and Data Mining Viewer Controls and add the reference to Microsoft.AnalysisService.Viewers.dll files, also in toolbox so i can drag Data Mining Viewer Controls on my Windows form.
I have a problem how to load data into Viewers (for example TreeViewer ...

I am trying to use run the statement below and Am getting the error:
##
Parser: The statement dialect could not be resolved due to ambiguity.
##

Is it possible to do Data Mining in SQL Server 2000, I have seen demos for SQL Server 2005.
If it is possible in SQL Server 2000, can anyone post resource links for the same.
thanks
Kiran

Is there a way to use PredictProbability([Admits_NN].[Mths], >6)?
Or will take only one state?
Thanks,
Andrew

Hi again!
I have permission problems in Analysis services using DMX statement.
In Analysis services I have two databases, DB1 and DB2. Both have Imersonation default. And each database has one data source, first database has data source DS1 and second has DS2. Both of two ...

To all,
How would or would not affect a model if you manually dicretize the data before running thru the algorithm or let the algorithm dicretize for you.
Does it make any difference if I set I.e:
Age: 60-70, 80-90, >90