Classifier - Software Development Case Studies in Java - page 60

root : Node

children children

label = “AC”

: Arc : Arc

label = “ABS”

destination destination

n1 : Node n1 : Node

children children children children

label =
“yes”

label =
“no”

: Arc : Arc : Arc : Arc

destination destination destination destination

l1 : Node l2 : Node l3 : Node l4 : Node

label = “high” label = “medium” label = “low”

Figure 4.7 Static structure of a sample decision tree

Using this class structure, the test decision tree, which is shown in
Figure 4.1, can be represented by a set of linked objects as shown in the
object diagram depicted in Figure 4.7.
A Node object, called root in the figure, whose label is “AC”, represents
the root node. Since the links from a node to its children have been objecti-
fied into Arc objects, the root node is connected to its children, n1 and n2,
through two Arc objects that have labels “yes” and “no” respectively. The
other nodes and arcs of the test tree are represented likewise using the class
structure shown in Figure 4.7.

4.4.3 Implementation

public class FeatureType {
private String[] allowedValues;
public FeatureType(String[] allowedValues) {
this .allowedValues # allowedValues;

}
public boolean isAllowed(String value) {
for ( int i # 0; i<allowedValues.length; !! i){