Scribd
Upload
79 views3 pages

WEKA Classification for Diabetes Data

The document describes an experiment using the WEKA data mining tool to perform classification on a diabetes dataset. The dataset contains attributes like glucose concentration, blood pressure, skin thickness, and age for 30 individuals, along with a class label of either tested positive or negative. WEKA's J48 decision tree algorithm was able to correctly classify 22 of 30 instances, with an accuracy of 75.86%. The decision tree used the age, glucose concentration, blood pressure, and skin thickness attributes to determine whether an individual tested positive or negative for diabetes.

Uploaded by

Shivam Dhar
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
79 views3 pages

WEKA Classification for Diabetes Data

The document describes an experiment using the WEKA data mining tool to perform classification on a diabetes dataset. The dataset contains attributes like glucose concentration, blood pressure, skin thickness, and age for 30 individuals, along with a class label of either tested positive or negative. WEKA's J48 decision tree algorithm was able to correctly classify 22 of 30 instances, with an accuracy of 75.86%. The decision tree used the age, glucose concentration, blood pressure, and skin thickness attributes to determine whether an individual tested positive or negative for diabetes.

Uploaded by

Shivam Dhar
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
You are on page 1/ 3

EXPERIMENT 2

AIM: Implementation of Classification technique on ARFF files using WEKA

Data Set

@relation diabetes1
@attribute 'glucose_cocentration' real
@attribute 'blood_pressure' real
@attribute 'skin_thickness' real
@attribute 'age' real
@attribute class {tested_negative,tested_positive}
@data
7,140,50,60,tested_positive
8,120,43,75,tested_positive
6,110,37,60,tested_positive
9,125,28,46,tested_positive
9,160,53,72,tested_positive
7,105,48,65,tested_positive
8,100,32,78,tested_positive
9,110,25,43,tested_positive
3,145,49,54,tested_positive
4,156,60,40,tested_positive
2,170,28,37,tested_positive
1,20,43,60,tested_positive
4,108,46,45,tested_positive
3,90,27,70,tested_positive
2,100,36,30,tested_negative
4,125,35,45,tested_negative
3,120,26,48,tested_negative
1,100,30,30,tested_negative
4,108,23,43,tested_negative
3,140,39,58,tested_positive
1,120,55,40,tested_positive
2,87,29,34,tested_negative
3,96,32,71,tested_positive
1,104,54,27,tested_negative
4,120,31,31,tested_negative
1,105,30,40,tested_negative
3,110,21,46,tested_negative
2,102,36,33,tested_negative
3,118,27,40,tested_negative
Result

=== Predictions on user test set ===

inst# actual predicted error prediction


1 1:? 2:tested_positive 1
2 1:? 2:tested_positive 1
3 1:? 2:tested_positive 1
4 1:? 2:tested_positive 1
5 1:? 2:tested_positive 1
6 1:? 2:tested_positive 1
7 1:? 2:tested_positive 1
8 1:? 2:tested_positive 1
9 1:? 2:tested_positive 1
10 1:? 2:tested_positive 1
11 1:? 2:tested_positive 1
12 1:? 2:tested_positive 1
13 1:? 2:tested_positive 0.667
14 1:? 2:tested_positive 1
15 1:? 1:tested_negative 1
16 1:? 1:tested_negative 1
17 1:? 1:tested_negative 1
18 1:? 1:tested_negative 1
19 1:? 1:tested_negative 1
20 1:? 2:tested_positive 1
21 1:? 2:tested_positive 0.667

Correctly Classified Instances 22 75.8621 %


Incorrectly Classified Instances 7 24.1379 %

=== Confusion Matrix ===

a b <-- classified as
10 2 | a = tested_negative
5 12 | b = tested_positive
Analysis:

At root node dataset is split into age attribute ,if age is >48 person is tested as positive
else next split is on basis of glucose_concentration .If glucose_concentration is >6
person is tested as positive else next split is blood_pressure. If blood_pressure >140
person is tested as positive else next split is skin_thickness. If skin_thickness >39 person
is tested as positive else person is tested as negative.

You might also like