Introduction to Decision Trees - Data Mining with Decision Trees: Theory and Applications - page 13

YRSJOB

<3

3.5 ≥

PAYINC

I. Rate

20%

<20%

≥

P

N

P

DEPEND

N

>0

=0

P N

Fig. 1.5 Actual behavior of customer.

“decision” nodes). In a decision tree, each internal node splits the instance
space into two or more sub-spaces according to a certain discrete function
of the input attributes values. In the simplest and most frequent case, each
test considers a single attribute, such that the instance space is partitioned
according to the attributes value. In the case of numeric attributes, the
condition refers to a range.
Each leaf is assigned to one class representing the most appropriate
target value. Alternatively, the leaf may hold a probability vector (anity
vector) indicating the probability of the target attribute having a certain
value. Figure 1.6 describes another example of a decision tree that predicts
whether or not a potential customer will respond to a direct mailing.
Internal nodes are represented as circles, whereas leaves are denoted as
triangles. Two or more branches may grow out from each internal node.
Each node corresponds with a certain characteristic and the branches
correspond with a range of values. These ranges of values must be
mutually exclusive and complete. These two properties of disjointness and
completeness are important since they ensure that each data instance is
mapped to one instance.
Instances are classified by navigating them from the root of the tree
down to a leaf according to the outcome of the tests along the path. We
start with a root of a tree; we consider the characteristic that corresponds