CS4341 Introduction to Artificial Intelligence
About Weka - C 2002
About the Weka System:
To experiment with the machine learning methods that we'll
study in class, you are encouraged to use the
Weka system
(http://www.cs.waikato.ac.nz/ml/weka/).
Weka is an excellent machine-leaning/data-mining environment.
It provides a large collection of Java-based mining algorithms,
data preprocessing filters, and experimentation capabilities.
Weka is open source software issued under the GNU General Public License.
Downloading the Weka System:
For more information on the Weka sytem, to download the system and
to get its documentation, look at
Weka's webpage
(http://www.cs.waikato.ac.nz/ml/weka/).
You should download and use the 3.2.1 GUI version of the system.
Input to the Weka System:
Input for Weka should be written in the Attribute-Relation File Format (ARFF).
For instance, the input data for the PlayTennis problem (see your decision
trees handouts) that we discussed
in class would be written as follows:
name of the file: weather.nominal.arff
@relation weather.symbolic
@attribute outlook {sunny, overcast, rainy}
@attribute temperature {hot, mild, cool}
@attribute humidity {high, normal}
@attribute windy {TRUE, FALSE}
@attribute play {yes, no}
@data
sunny,hot,high,FALSE,no
sunny,hot,high,TRUE,no
overcast,hot,high,FALSE,yes
rainy,mild,high,FALSE,yes
rainy,cool,normal,FALSE,yes
rainy,cool,normal,TRUE,no
overcast,cool,normal,TRUE,yes
sunny,mild,high,FALSE,no
sunny,cool,normal,FALSE,yes
rainy,mild,normal,FALSE,yes
sunny,mild,normal,TRUE,yes
overcast,mild,high,TRUE,yes
overcast,hot,normal,FALSE,yes
rainy,mild,high,TRUE,no
This file is available with the system (look under the
"data" directory once that you download the Weka system).
You are encouraged to play with the system running classifiers
over this dataset.
Machine Learning Methods in the Weka System:
Weka provides a large variety of machine learning techniques.
Among them, decision trees and neural networks.
- Decision Trees Run the ID3 algorithm (available under "Classifiers")
Remarks
We'll post more on the Weka system as needed on this webpage.
Wendy Kogel and Greg Miletter (two your TAs) are very familiar
with the Weka System. Use their expertise if you have any questions.