well, i;m doing a thesis about data mining right now.And i found some difficulties. In my thesis i use visual basic 6.0 and sql server 2000 database and sql server 2000 analysis service. For information i use DSO (Decision Support Objects in programming). I've read claude seidmann book about data mining and i really help with that book. But i still have some questions about data mining using DSO :
1. I have made a relational mining model using VB. I've created the code based on Claude Seidman book. well it's process well. But when i want to browse the result in analysis manager there's an error message ;
"Unable to browse the mining model, object or provider is not capable of performing requested operation"
I really don't understand with this error message. Could you tell me what's the meaning of this error message?
2. shoud i make another code in VB to browse the result ?
3. I still don't understand about Microsoft decision tree algorithm works.So could you tell me how microsoft decision tree wroks?
Thank you so much for your attention and i really appreciate with your help.Thank you
Here is some information on Decision Trees algorithm from SQL SERVER 2005 perspective.
http://msdn2.microsoft.com/en-us/library/ms175312(SQL.90).aspx
HTH..
To do a data mining project, may I recommend using Sql server 2005 and Visual Studio 2005? You could download free trial versions from Microsoft site.
ZULFIQAR SYED
HTTP://ZULFIQAR.TYPEPAD.COM
|||It's likely that some of the client components in your Analysis Services 2000 install are not registered correctly. Re-installing the latest service pack (SP4) should fix the problem you're seeing.
Zulfiqar's reply has valid suggestions about trying SQL Server 2005 Data Mining if you're starting a new data mining project and checking the SQL 2005 Books Online for information about the decision trees algorithm. More in-depth information about the algorithm can be found in the research papers accessible from this tip: http://www.sqlserverdatamining.com/DMCommunity/TipsNTricks/986.aspx
No comments:
Post a Comment