Call now: 252-767-6166  
Oracle Training Oracle Support Development Oracle Apps

 
 Home
 E-mail Us
 Oracle Articles
New Oracle Articles


 Oracle Training
 Oracle Tips

 Oracle Forum
 Class Catalog


 Remote DBA
 Oracle Tuning
 Emergency 911
 RAC Support
 Apps Support
 Analysis
 Design
 Implementation
 Oracle Support


 SQL Tuning
 Security

 Oracle UNIX
 Oracle Linux
 Monitoring
 Remote s
upport
 Remote plans
 Remote
services
 Application Server

 Applications
 Oracle Forms
 Oracle Portal
 App Upgrades
 SQL Server
 Oracle Concepts
 Software Support

 Remote S
upport  
 Development  

 Implementation


 Consulting Staff
 Consulting Prices
 Help Wanted!

 


 Oracle Posters
 Oracle Books

 Oracle Scripts
 Ion
 Excel-DB  

Don Burleson Blog 


 

 

 


 

 

 

 
 

Adding a Data Source to the Model

Data warehouse tips by Burleson Consulting

This is an excerpt from Dr. Ham's premier book "Oracle Data Mining: Mining Gold from your Warehouse".

Next, we select the apply data source that we created as MINING_DATA_BUILD_V_NOUS in Step 2 of the Apply Data wizard, and click Next. 

Now you have the option of selecting additional columns to be included in the table resulting from the Apply operation.  The wizard suggests that you include the customer identity attribute so that you can see which customers are most likely to have an affinity card. 

By default, the ?Apply Result? contains only the case identifier and prediction information.  You may want to keep the ?bare bones? set of predictor variables and join this with tables containing the customer contact information later.  Select some or all of the attributes in Step 3, and click Next. 

The next step allows you to choose the format for the output table.  When the model is applied to a particular customer, a score is generated for each possible target value.  A sorted list is generated from the most to least likely value.  This list will have only two entries since our target is a binary, because our customers either have or do not have an affinity card. 

Viewing Top Rankings

If your target had multiple results, as in the case of predicting which of 10 stores a customer was most like to shop, then you might want to see the ranking of top three choices for each customer, and would click the radio button next to ?Number of Best Target Values? and enter in 3.  The output table would have three rows for each individual containing the prediction information for the top three stores.

In the case of which of ten stores the customer is most likely to shop, you may only be interested in a customer?s ranking probability for a particular store.  Then you would check the radio button next to ?Specific Target Values? and check the box next to the particular store you were interested in.  The result would be a table with one row for each customer with the prediction for that store, even though the probability might be extremely low. 

Using the Classification Apply Option

The Classification Apply Optionhas as the default the most probable target value or lowest cost.  In our case, we?ll keep the default selection and click Next, opting for an output table with one row per customer.  Choose a name for the Apply Data Mining Activity, such as ALL_US_APPLY_NON_US.

When the Activity has finished running, click Result in the Apply section to view the output table.  Since we chose the ?Most Probable? classification option, the table contains the customer or case ID, then the predictor variables (prediction, probability, cost and rank), then the additional attributes you chose to have in the output table. 

The prediction is the most likely target value, and the probability is the confidence in that prediction.  Cost represents the cost of a wrong prediction, with low cost meaning high probability.  The rank will be one for all cases since we did not choose a predictor variable with more than 2 possible outcomes, and consequently do not have more values to predict. 

We have now completed building, testing and applying a model to our customer data.  We can go back to the boss having produced a mailing list in our Oracle data warehouse that should draw in the most promising customers.   
 

For more tips and tricks for Oracle data warehouse analysis, see Dr. Ham's premier book "Oracle Data Mining: Mining Gold from your Warehouse"

You can buy it direct from the publisher for 30%-off:

http://www.rampant-books.com/book_2006_1_oracle_data_mining.htm


 

 

��  
 
 
Oracle Training at Sea
 
 
 
 
oracle dba poster
 

 
Follow us on Twitter 
 
Oracle performance tuning software 
 
Oracle Linux poster
 
 
 

 

Burleson is the American Team

Note: This Oracle documentation was created as a support and Oracle training reference for use by our DBA performance tuning consulting professionals.  Feel free to ask questions on our Oracle forum.

Verify experience! Anyone considering using the services of an Oracle support expert should independently investigate their credentials and experience, and not rely on advertisements and self-proclaimed expertise. All legitimate Oracle experts publish their Oracle qualifications.

Errata?  Oracle technology is changing and we strive to update our BC Oracle support information.  If you find an error or have a suggestion for improving our content, we would appreciate your feedback.  Just  e-mail:  

and include the URL for the page.


                    









Burleson Consulting

The Oracle of Database Support

Oracle Performance Tuning

Remote DBA Services


 

Copyright © 1996 -  2017

All rights reserved by Burleson

Oracle ® is the registered trademark of Oracle Corporation.

Remote Emergency Support provided by Conversational