Oracle Data Mining Sample Views

The structure tab depicts all attributes of the view, including primary key (PK), column name (Name), format of the data (Type), size of the field (Size), Scale and whether NULL is allowed.   If any comments were added when the view was created, these will be seen here too. 

The ODM Data Tab

The data tab queries the view and shows data for 100 rows.  Fetch Next brings in an additional 100 rows, and so on for every time you click Fetch Next.  Refresh re-runs the query. 

Only views will have the View Lineage tab, and using this you can see where the data was derived.  

You can see that the MINING_DATA_BUILD_Vview was built using three tables from the SH schema in the Oracle database. 

Exploring Data Detail

Now that we’ve seen how the data was generated in the view, we’ll begin to explore the data in more detail.  Right click MINING_DATA_BUILD_V in the navigator window and select Show Summary Single-Record.  Because each customer is contained on only one row of the data set, we select Single-Record rather than Multi-Record.  

A new window opens up titled Data Summarization viewer.  Note that in this instance the sample count is 999 records.  Keep in mind that you can change the sample count by going to Tools, Preferences, and choosing the Sampling tab.  You might like to try this now, and change the setting to 2000.   Next time you view the Data Summarization viewer, you’ll notice that now the sample count is 1500, which is the entire data set.  If you are reporting the summary statistics shown in the viewer, you’ll sometimes want the entire data set.   If you need to report on the average age of your clientele, you can say that the average age is 38.89 years, ranging from 17 to 90 with a statistical variance of 185.95.