Scribd
Upload
125 views7 pages

Creating .arff Data-set for WEKA

The document describes creating a data set in the .arff file format for student records to demonstrate preprocessing in WEKA. It provides code to define attributes like student name, roll number, and subject grades. The data set is then opened in WEKA and different filters like normalization, discretization, and randomization are applied to preprocess the data.

Uploaded by

Student
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
125 views7 pages

Creating .arff Data-set for WEKA

The document describes creating a data set in the .arff file format for student records to demonstrate preprocessing in WEKA. It provides code to define attributes like student name, roll number, and subject grades. The data set is then opened in WEKA and different filters like normalization, discretization, and randomization are applied to preprocess the data.

Uploaded by

Student
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

PROGRAM NO-1

OBJECTIVE: To create data-set in .arff file format. Demonstration of


preprocessing on WEKA data-set.

Creating the Data-set in .arff file format for Student Record.


Wite the below code in notepad or any other editor.

@relation studentrecord

@attribute [Link]. numeric

@attribute Name{Ram,Shyam,Arjun,Kiran,Sama,Rama,Gaama,Beta,Alpha,Theeta}

@attribute RollNo. numeric

@attribute Physics numeric

@attribute Maths numeric

@attribute Chemistry numeric

@data

1,Ram,8,100,100,100

2,Shyam,7,98,99,96

3,Arjun,3,99,98,98

4,Kiran,4,96,97,99

5,Sama,5,95,96,100

6,Rama,6,94,96,95

7,Gaama,9,93,97,97

8,Beta,2,92,99,96
9,Alpha,1,91,94,98

10,Theeta,10,90,95,100

11,Rama,11,100,100,99

Save it in .arff format.

To Preprocess:
1-Open WEKA tool go to the Preprocess option on the top left corner.

2-Click on the Open file….

Choose the file that has been save at any location in your PC.

3-After opening it gives the overall data about the data that has been saved and
its various aspect as shown below.
4-Now filter can be applied to display out the value of required fields.

5-Choose the filter

i)Attribute Selection
ii)Normalise

Go to the filter as shown above and click to choose


weka>filter>unsupervised>Normalise.(Here one attribute is removed that is of
subject).

And in the attribute field choose rollno.

Result obtained is:


iii)Discretize

Go to the filter as shown above and click to choose


weka>filter>unsupervised>Discretize.(Here three field from subject is taken-
physics,maths,chemistry).

Result Obtained:
iv)Randomize

Go to the filter as shown above and click to choose


weka>filter>unsupervised>Randomize.(Here two field from subject is taken-
maths,chemistry).

You might also like