Building Data Mining Tools Using JDM - Java Data Mining: Strategy, Standard, and Practice - page 346

The Check button imports the metadata on the dataset specified by
the URI. JDM does not fully specify the results of import metadata,
but the list of dataset attributes should be provided at that time.
Some vendors may also compute some basic attribute statistics. This
is why, when the user selects “Check,” the drop-down lists containing
the possible attribute names are updated. For example, on the screen
shown in Figure 13-3, the drop-down list allowing the selection of the
key attribute, and the one used to select a target attribute have been
updated with the list of columns contained in the dataset referenced
by the URI “file:///./Data/Census.csv.”
The Stats button is only active when the vendor implementation
supports the ComputeStatisticsTask interface. This allows quick data
exploration, providing univariate statistics that are displayed in a
pop-up window (described in Section 13.3.5).
Note that while the user specifies only the model name, this appli-
cation uses this name as a base name from which to generate names
of other persisted objects, such as tasks, build settings, and physical
datasets. We will see that automated name creation must be validated
against the maximum name length supported by the DME, assuming
that uniqueness is guaranteed by the user-provided names.
The source code is contained in the file JDMBuildConsole.java.
As explained for the administration console, once a connection is
obtained, the application invokes the start method. Following is the
code for the BuildConsole :

1. protected void start(Connection iConnection) {
2. mDmeConn iConnection;

3. try {

4. setDatasetSettingsEnabled(true);

5. MiningFunction[] lFunctions mDmeConn.getSupportedFunctions();

6. List lFunctionList Arrays.asList(lFunctions);

7. mFunctionPane.setEnabledAt(ATTRINDEX,

8. lFunctionList.contains(MiningFunction.attributeImportance));

9. mFunctionPane.setEnabledAt(CLASSIFINDEX,

10. lFunctionList.contains(MiningFunction.classification));

11. mFunctionPane.setEnabledAt(CLUSTERINGINDEX,

12. lFunctionList.contains(MiningFunction.clustering));

13. mFunctionPane.setEnabledAt(REGINDEX,

14. lFunctionList.contains(MiningFunction.regression));

15. mFunctionPane.setEnabledAt(ASSOCINDEX,

16. lFunctionList.contains(MiningFunction.association));

17. handlerCheckDataSet();

18. } catch (JDMException lException) {