Regarding DataMining in SQL Server 2005 Analysis Services

Hi to all

I have a complex scenario , so first I want to ask out the feasibility of it . I think its better if I state the scenario and some one on this forum reply to it , I need to build an application (e.g credit card application , Loan Application etc ) that requires some approval from expert , what I firstly want is that I apply datamining on this data so that next time when I enter the data the result (approval or reject ) should be given by datamining tool , this I gather is poosible by using Analysis Services in SQL 2005 , but I also want the bases of that decision ( I mean the rules/some thing else that the Datamining created agaisnt the data entered ) , So can any one do any help on this , you can also reach me at razi_rais@.yahoo.com . its preety much urgent so your prompt response is higly appreciated.

Thanks and Regards

Razi Bin Rais

Moving to DM forum

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

|||

Data Mining definately has the ability to do this. Since you are apparently new at data mining I would recommend the Data Mining with SQL Server 2005 book and also using the Data Mining Addins for Office 2007 to get started in learning and exploring the tools.

Instructions for getting the preview build of the addins as well as eval editions of Office and SQLServer are here: http://blogs.msdn.com/jamiemac/archive/2006/12/29/try-out-the-data-mining-addins-for-office-2007-free-for-60-days.aspx

|||

Hi Jamie

Thanks for your feedback , I will certainly look into resources you specify . For now can you please amplify on two points

1) Is DM in SQL Server 2005 provides any way to get the rules out of it , on basics of which it makes decision. I mean through object model , or plain xml ... anything?

2) Can you give me simple example of application (step wise / demograpic / architecutal ) like say credit card application or loan approval application that has very simple schema and only have accept or reject option apart from simple personal details of customer that will be predicted by DM (After entering some sampel data ) , and if possible the above point 1 for it .

Thanks again

Razi

|||

This link might be helpful: http://www.sql-server-performance.com/ec_data_mining.asp

It includes an high-level example of a credit approval application utilizing data mining...

Frank

sql

Regarding Bulk Copy in Transactional Replication..

Hi,
I got typical scenario in transacrtional replication. i want to avoid bulk
copy during replication.
Let me explain my scenario.
I’m doing transactional replication between two databases.
When publisher and subscriber created the data going to be bulk copied from
publisher table to subscriber table.
My main intension was to create replication between different tables with
different fields in which I got succeeded.
But main problem is I want to stop this bulk copy from publisher to
subscriber.
Scenario 1: my subscriber table may contain some previous data which will be
replaced with publisher data due to bulk copy.
I don’t want this .I want to avoid this bulk copy and wants to create
procedures(for insert, update and delete transactions) in subscriber which
will take care of replication.
I achieved almost everything but not able to avoid this bulk copy during
the creation of subscriber.
As I know the only way I can stop bulk copy is by creating subscription
without subscription agent. But here without subscription agent the
procedures(for insert, update and delete transactions)
won’t get created in subscriber.
Help me regarding the above scenario and I need it urgently.
Regards
Baji Prasad
I'm not entirely clear about your scenario, but it seems to me that you are
talking about nosync initialisations. Please have a look here and see if this
applies:
http://www.replicationanswers.com/NoSyncInitializations.asp
HTH,
Paul Ibison
|||hi paul..
im very thankfull for ur reply.
yes it helped me a lot atleast i got an idea how to avoid bulk copy.
again thanks man...
Regards
Baji Prasad
"Paul Ibison" wrote:

> I'm not entirely clear about your scenario, but it seems to me that you are
> talking about nosync initialisations. Please have a look here and see if this
> applies:
> http://www.replicationanswers.com/NoSyncInitializations.asp
> HTH,
> Paul Ibison
>
sql

Regarding Bulk Copy During Transactional Replication

Hi ,

I got a problem in regarding Transactional Replication.

Let me explain my scenario.

I’m doing transactional replication between two databases.

When publisher and subscriber created the data going to be bulk copied from publisher table to subscriber table.

My main intension was to create replication between different tables with different fields in which I got succeeded.

But main problem is I want to stop this bulk copy from publisher to subscriber.

Scenario 1: my subscriber table may contain some previous data which will be replaced with publisher data due to bulk copy.

I don’t want this .I want to avoid this bulk copy and wants to create procedures(for insert, update and delete transactions) in subscriber which will take care of replication.

I achieved almost everything but not able to avoid this bulk copy during the creation of subscriber.

As I know the only way I can stop bulk copy is by creating subscription without subscription agent. But here without subscription agent the procedures(for insert, update and delete transactions)

won’t get created in subscriber.

Help me regarding the above scenario and I need it urgently.

in sp_addsubscription, look at paramter @.sync_type, you can specify 'replication support only' and 'initialize with backup'. That allows you to skip the bcp. You can find more information about the latter here: http://msdn2.microsoft.com/de-de/library/ms147834(SQL.90).aspx.

Regarding Bulk Copy During Transactional Replication

Hi ,

I got a problem in regarding Transactional Replication.

Let me explain my scenario.

I’m doing transactional replication between two databases.

When publisher and subscriber created the data going to be bulk copied from publisher table to subscriber table.

My main intension was to create replication between different tables with different fields in which I got succeeded.

But main problem is I want to stop this bulk copy from publisher to subscriber.

Scenario 1: my subscriber table may contain some previous data which will be replaced with publisher data due to bulk copy.

I don’t want this .I want to avoid this bulk copy and wants to create procedures(for insert, update and delete transactions) in subscriber which will take care of replication.

I achieved almost everything but not able to avoid this bulk copy during the creation of subscriber.

As I know the only way I can stop bulk copy is by creating subscription without subscription agent. But here without subscription agent the procedures(for insert, update and delete transactions)

won’t get created in subscriber.

Help me regarding the above scenario and I need it urgently.

in sp_addsubscription, look at paramter @.sync_type, you can specify 'replication support only' and 'initialize with backup'. That allows you to skip the bcp. You can find more information about the latter here: http://msdn2.microsoft.com/de-de/library/ms147834(SQL.90).aspx.

Refresh a table

Can i referesh a open table?

The scenario is that i have two windows open... one is the table and the other is the query that modifies the same table. I was wondering if there is a way to refresh the table so that after i modify the table i don't have to close it and reopen it to verify the changes.

cheers

In open table, you can refresh by clicking on the "Execute" button, the red exclamation point.

Peter

sql

Referencing the temporary Dataset in Lookup control SQL

Does anyone know how to reference the ongoing dataset within the dataflow in sql?

So here is the scenario I have.

1) OLE DB SOURCE
2) Lookup a value and that column gets added to the dataset
3) Lookup another value but this time I would rather code the sql rather than select a referencing table. How do I reference the current dataset?

Thanks,
ChestonI'm not clear on what you are trying to do.

You can use SQL in a lookup transformation. You can even build a lookup table with literal values in SQL in that lookup transformation without hitting a table.|||Let me see if I can be more clear on what I am trying to do.

Table 1 - Ole DB source control
Lookup 1 - Reference another table and based on keys add a foreign key
Lookup 2 - I wish to perform a sql statement to update the newly added foreign key from lookup 1 to a default value if it is null. How do I reference the dataset?

So in the Lookup 2 I select SQL query:
Update dataset set foreign key = 0 where foreign key is null

Then continue on with the dataset with more lookups/transformations.

Does this help?

Cheston|||Use a derived column transformation to set that column to 0 if it is null.