Scribd
Upload
13 views

AIML Lect5 Assignment ID3

1. The document describes applying the ID3 algorithm to classify records in a dataset by position using attributes like department, age, and salary. 2. It constructs a decision tree with age as the root node, then salary as the next node for records with age 31-40, and finally department as the last node. 3. The decision tree can classify new records by determining their position based on traversing the tree using their attribute values for age, salary, and department.

Uploaded by

Yash Thakur
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
13 views

AIML Lect5 Assignment ID3

1. The document describes applying the ID3 algorithm to classify records in a dataset by position using attributes like department, age, and salary. 2. It constructs a decision tree with age as the root node, then salary as the next node for records with age 31-40, and finally department as the last node. 3. The decision tree can classify new records by determining their position based on traversing the tree using their attribute values for age, salary, and department.

Uploaded by

Yash Thakur
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

1. Consider a dataset consisting of 11 records in table 1 of an organization “XYZ”.

Consider Position
attribute as a class for classification. Construct full decision tree by applying ID3 algorithm.
Department Age Salary Position
Personnel 31-40 Medium Range Boss
Personnel 21-30 Low Range Assistant
Personnel 31-40 Low Range Assistant
MIS 21-30 Medium Range Assistant
MIS 31-40 High Range Boss
MIS 21-30 Medium Range Assistant
MIS 41-50 High Range Boss
Administration 31-40 Medium Range Boss
Administration 31-40 Medium Range Assistant
Security 41-50 Medium Range Boss
Security 21-30 Low Range Assistant
Table-1

We are applying ID3 algorithm, on the above dataset as follows:


The initial entropy = (6/11)log(11/6) + (5/11)log(11/5) = 0.29923
(Assistant) (Boss)

Gain(Department)= 0.29923- { (3/11)[(1/3)log3 +(2/3)log(3/2)] + (4/11)[ (2/4)log 2 + (2/4)log 2] +


(2/11)[ (1/2)log 2 + (1/2)log 2 ] + (2/11)[ (1/2)log 2 + (1/2)log 2 ] }
= 0.29923- 0.2943
= 0.0049

Gain(Age) = 0.29923- { (4/11)[(4/4)log(4/4)] + (5/11)[(3/5)log(5/3) + (2/5)log (5/2)] +


(2/11)[(2/2) log (2/2)] }
= 0.29923 -0.1328
=0.1664

Gain(Salary) = 0.29923 – { (3/11)[(3/3)log 3] + (6/11)[(3/6) log2 + (3/6)log 2] +


(2/11) [(2/2 log(2/2) ) }
= 0.29923 – 0.164
= 0.1350

Since Age is having the maximum gain so this attribute is selected as the first splitting attribute. In age
range 31-40, class is not defined while for others range it is defined. So we have to again find out the
spitting attribute for this age range (31-40). Now the tuples that belong to this range is as follows:
Department Salary Position
Personnel Medium Range Boss
Personnel Low Range Assistant
MIS High Range Boss
Administration Medium Range Boss
Administration Medium Range Assistant

Again the initial entropy = (2/5)log(5/2) + (3/5)log(5/3) = 0.29922


(Assistant) (Boss)
Gain(Department)= 0.29922- { (2/5)[ (1/2)log 2 + (1/2)log 2 ] +1/5[ (1/1)log1] +
(2/5)[ (1/2)log 2 + (1/2)log 2 ] }
=0.29922- 0.240
=0.05922

Gain(Salary) = 0.29922- { (1/5) [(1/1)log1 ] +(3/5)[ (1/3)log 3 + (2/3)log(3/2)] +


(1/5) [(1/1)log1 ] }
= 0.29922- 0.1658
=0.13335

The Gain is maximum for salary attribute, so we take salary as the next splitting attribute. In salary middle
range, class is not defined while for others range it is defined. So we have to again find out the spitting
attribute for this middle range. Since only department is left so the department will be next splitting
attribute. Now the tuples that belong to this salary range is as follows:

Department Position
Personnel Boss
Administration Boss
Administration Assistant

Again in Personnel department all persons are Boss while in the Administration there is tie between the
classes. So, the person can be either Boss or Assistant in Administration department.
Now the decision tree is as follows:

Age ?

21-30 41-50
31-40

Assistant
Boss

Salary ?

High Range
Low Range
Medium range

Assistant Boss

Department ?

Administration
Personnel

Boss
Assistant/Boss

Now we will take a new dataset and we will classify the class of each tuple by applying the decision tree
that we have build above.

You might also like