Using WEKA with Java applications
Step 1:
Launch Weka by clicking on: RunWeka.bat
Select ‘Explorer' from the list of Weka Applications.
Select the ‘Preprocess' tab and click on ‘Open File'. Choose the file ‘Sunburn.arff'.
Review the contents of the file. Weka gives a summary of the attributes and their distribution.
Select the ‘Classify' tab and make sure the following Classifier (J48) and Test options (Use training set) settings are selected.
Press ‘Start' to create a classification model from the ARFF file.
Details about the classifier are contained in ‘Classifier output'.
Right click on ‘trees J48' in the ‘Result List' and select ‘Visualise Tree'.
A tree representation of the learning model will be displayed.
Q. What is the defining feature of the decision tree?
Q. Which features seem to be redundant?
Step 2:
In notepad, open file ‘sunburn2.arff'
Add an additional attribute ‘shade' to the list of attributes:
@ATTRIBUTE 'shade' {yes, no}
The values for ‘shade' should be listed at the start of each vector. For instance the first vector:
blonde, average,light, no, burned
becomes:
no,blonde, average,light, no, burned
Values (in order, top to bottom) for each vector are as follows:
no, no, no, no, no, no, no, no, no, no, no, yes, yes, no, no, no
Accordingly, update each vector in the file ‘sunburn2.arff' and choose save.
Step 3:
In WEKA Explorer click the ‘Preprocess' tab and then click ‘Open File'. Select the new file ‘sunburn2.arff' and repeat Step 1 to create a new J48 decision tree with this file.
Q. Does the classification accuracy increase or decrease for this new file?
Q. Does the J48 decision tree change, if so in what way?
Q. What becomes the defining feature of the decision tree?
Step 4:
In WEKA Explorer stay in the ‘Classify tab.
Select the ‘Supplied Test set'radio button and click the ‘Set' button, following by the ‘Open file' button. Choose the file ‘sunburn2TEST.arff' and click ‘Close'.
Click the ‘More Options' button and ensure there is a tick beside ‘Output predications' then press ‘OK'.
Right click on ‘tree J48' and select ‘Re-evaluate model on current test set'.
The prediction results will appear in the ‘Classifer output' under the heading ‘Predictions on test set'.
Compare the predictions to the feature vectors in the file ‘sunburn2TEST.arff'. Are the predictions reasonable? Are they as you would expect?
Step 5:
Open the folder: Apps\Weka\Data
Choose one of the .arff files and repeat Step 1.
Q. What is the defining feature of the decision tree?
Q. Which features seem to be redundant?
Attachment:- Practical4.rar