WALTER HUDSON's ELECTRONIC JOURNAL
DATA MINING

HOME

CH 1
CH 2
CH 3
CH. 4
CH. 5
CH. 6
CH. 7
CH.8
CH. 9
CH.10
CH. 11
D-S-S
OLAP
DATA MINING
Smart Card
MSSP
LOUD CLOUD
CASE STUDY: THE VALUE OF IS
REAL WORLD CASE
amazon.com
MRS. FIELD'S COOKIES
Application Exercise 3.3
GROUP PROJECT(PAPER)
ZD NET ARTICLE
Real World Case #3

 

TABLE  1

Cases

Age

Income

Net Assets

Credit History

 

Outcome

1

21

8,000

-1500

no

Default

2

36

38000

8000

good

on-time

3

29

26000

12000

poor

Late

4

19

6500

2000

no

Late

5

42

55000

46000

good

on-time

6

64

22000

36000

poor

Default

7

28

36000

26000

poor

Late

8

42

42000

66000

good

on-time

9

35

22000

-4000

good

on-time

10

48

33500

6000

good

on-time

11

26

16000

-8000

poor

default

12

28

17000

2000

no

late

13

17

7000

1000

no

late

14

24

55000

40000

good

on-time

15

29

26000

8000

no

late

16

19

11000

1500

good

on-time

17

26

7000

3500

no

late

18

47

22000

-6000

poor

default

A

31

27,000

-5,000

good

?

B

27

45,000

-12,000

no

?

 

 

TABLE  2

Cases

Age

Income

Net Assets

Credit History

 

Outcome

 

A

 

B

1

<25

<10,000

<0

no

default

 

 

2

>25

25,000-40,000

0-10,000

good

on-time

3

1

3

>25

25,000-40,000

>10,000

poor

late

 

 

4

<25

>40,000

0-10,000

no

late

 

 

5

>25

>40,000

>10,000

good

on-time

 

 

6

>25

10,000-25,000

>10,000

poor

default

 

 

7

>25

25,000-40,000

>10,000

poor

late

 

 

8

>25

> 40,000

>10,000

good

on-time

 

 

9

>25

10,000-25,000

<0

good

on-time

 

 

10

>25

25,000-40,000

0-10,000

good

on-time

 

 

11

>25

10,000-25,000

<0

poor

default

 

 

12

>25

10,000-25,000

0-10,000

no

late

 

 

13

<25

<10,000

0-10,000

no

late

 

 

14

<25

> 40,000

>10,000

good

on-time

 

 

15

>25

25,000-40,000

0-10,000

no

late

 

 

16

<25

10,000-25,000

0-10,000

good

on-time

 

 

17

>25

<10,000

0-10,000

no

late

 

 

18

>25

10,000-25,000

<0

poor

default

 

 

A

>25

25,000-40,000

<0

good

-

-

-

B

>25

> 40,000

<0

no

-

-

-

 

 

TABLE  3

Cases

Age

Income

Net Assets

Credit History

Outcome

max

75

100,000

100,000

1

-

min

16

0

-50,000

0

-

1

0.085

0.080

0.323

0

3

2

.338

.38

.386

0

1

3

.220

.26

.413

0

2

4

.050

.065

.346

0

2

5

.440

.55

.64

1

1

6

.814

.22

.573

0

3

7

.203

.36

.506

0

2

8

.440

.42

.773

1

1

9

.322

.22

.306

1

1

10

.542

.335

.373

1

1

11

.169

.16

.28

0

3

12

.203

.17

.346

0

2

13

.016

.070

.34

0

2

14

.135

.55

.6

1

1

15

.220

.26

.386

0

2

16

.050

.11

.343

1

1

17

.169

.070

.356

0

2

18

.525

.22

.293

0

3

A

.254

.27

.3

1

-

B

.186

.45

.253

0

-

 

 

 

 

 

 

 

 

TABLE  4

Outcomes

Age

Income

Net Assets

Credit History

A

B

on-time

0.325

0.366

.488

1

0.356

1.291

late

0.154

0.179

0.384

2

1.260 

.44 

default

0.398

0.17

0.367

3

 1.320

 .610

 

 

 

 

TABLE 5

Rule 1

Credit History = good

Pay on-time

0.889

Rule 2

Credit History = no

Pay late

0.750

Rule 3

Credit History = poor

Income > $22,000

Pay late

0.750

Rule 4

Credit History = poor

Income <=$22,000

Default

0.800

 

 

 I BELIEVE "a" WILL PAY ONTIME.

 

I BELIEVE "b" WILL BE LATE.