Java Data Mining Concepts - Java Data Mining: Strategy, Standard, and Practice - page 148

5

100

90

4

80

70

3 60

50

40

2

30

20

1

10

0

0

0123456789 0

0123456789 0

Quantile Number Quantile Number

(a) (b)

With Attrition Model A

With Attrition Model B Without Model (Random)

Figure 7-8 Lift charts: (a) Cumulative lift, (b) cumulative gain.

Following are the typical high-level steps used to compute cumula-
tive lift and gain values:

• Compute the positive target value probability for all test
dataset cases.
• Sort cases in descending order of the positive target value
probability.
• Split the sorted test dataset cases into n groups, also known
as quantiles .
• Compute lift for each quantile—the ratio of the cumulative
number of positive targets and cumulative number of posi-
tive targets that can be found at random.
• Compute the cumulative gain for each quantile—the ratio
of cumulative predicted number of positive targets using
the model and total number of positive targets in the test
dataset.

Table 7-5 details lift and cumulative gain calculations for our cus-
tomer dataset example. Each row of this table represents a quantile
that has 100 customer cases. In the test dataset, there are 1,000 cases,
of which 190 are known to be attriters. Hence, picking a random
customer from this dataset, we have a 19 percent probability that the
customer is an attriter.