Data Science

 Classification: Basic Concepts

Decision Tree Induction

Bayes Classification Methods

Rule-Based Classification   

Assignment 4

1- Consider the data in the following table: 

  

TID 

Home Owner

Marital Status

Annual Income

Defaulted   Borrower

 

1

Yes

Single

[120 – < 150K]

No

 

2

No

Married

[90 – < 120K]

No

 

3

No

Single

[60 – < 90K]

No

 

4

Yes

Married

[120 – < 150K]

No

 

5

No

Divorced

[90 – < 120K]

Yes

 

6

No

Married

[60 – < 90K]

No

 

7

Yes

Divorced

[120 – < 150K]

No

 

8

No

Single

[90 – < 120K]

Yes

 

9

No

Married

[60 – < 90K]

No

 

10

No

Single

[90 – < 120K]

Yes

Let Defaulted Borrower be the class label attribute. 

a) Given a data tuple X = (Home Owner= No, Marital Status= Married, Income= $120K). What would a naive Bayesian classification of the Defaulted Borrower for the tuple be?

2- Consider the training example in the following table for a binary classification problem.

  

Customer ID

Gender

Car Type

Shirt Size

Class

 

1

M

Family

S

C0

 

2

M

Sports

M

C0

 

3

M

Sports

M

C0

 

4

M

Sports

L

C0

 

5

M

Sports

XL

C0

 

6

M

Sports

XL

C0

 

7

F

Sports

S

C0

 

8

F

Sports

S

C0

 

9

F

Sports

M

C0

 

10

F

Luxury

L

C0

 

11

M

Family

L

C1

 

12

M

Family

XL

C1

 

13

M

Family

M

C1

 

14

M

Luxury

XL

C1

 

15

F

Luxury

S

C1

 

16

F

Luxury

S

C1

 

17

F

Luxury

M

C1

 

18

F

Luxury

M

C1

 

19

F

Luxury

M

C1

 

20

F

Luxury

L

C1

a) Find the gain for Gender, Car Type, and Shirt Size.

b) Which attribute will be selected as the splitting attribute? 

Tags: No tags