Java Reference
In-Depth Information
3.
new DefaultComboBoxModel(iAttributesNames));
4.
mTargetNameClassif.setModel(
5.
new DefaultComboBoxModel(iAttributesNames));
6.
mTargetNameRegression.setModel(
7.
new DefaultComboBoxModel(iAttributesNames));
8.
mKeyForAssociation.setModel(
9.
new DefaultComboBoxModel(iAttributesNames));
10.}
The reader should now be familiar with the notion of
ComboBox-
Model,
which contains the data underlying the graphical compo-
nents. The combo box updated at line 2 enables selecting the
attribute that will be used as a target for the attribute importance
mining functions. The one updated at line 4 is for the target of the
classification function, and the one updated at line 6 is for the regres-
sion function. The one at line 8 allows the user to select a key (unique
identifier) attribute. This is required for association rules, but the key
attribute will also be forwarded to all build settings objects.
13.3.3
Computing Statistics
When the user clicks on the
Stats
button of the GUI, the following
code is executed:
1. private void handlerRetrieveStatistics() {
2.
try {
3.
boolean lDoesSupport
mDmeConn.supportsCapability(null,
4.
5.
null,
6.
MiningTask.computeStatisticsTask);
7.
if (!lDoesSupport) {
8.
displayLongMessage(this, "Unsupported feature",
9.
"This implementation does not support "
"the compute statistics task",
10.
11.
JOptionPane.INFORMATION_MESSAGE);
12.
return;
13.
}
PhysicalDataSet lStatData
retrievePhysicalDataset();
14.
if (lStatData
null) {
15.
16.
return;
17.
}
JDialog lDialog
new JDialog(this,
18.
19.
"Descriptive Statistics for URI:"
lStatData.getURI());
20.
21.
lDialog.getContentPane().add(
22.
new DescriptiveStatisticDisplayer(lStatData));
Search WWH ::
Custom Search