In this homework you will compare decision trees, neural networks and Naive Bayes on two different learning tasks. For decision trees, you will be using the C4.5 algorithm developed by Ross Quinlan. A great page describing the procedure for installing and running C4.5 can be found here.
For feed-forward neural networks, there are a number of software packages available. I would recommend Billnet. A local copy of the package and documentation can be found here.
For naive Bayes, probably the easiest option would be to code it yourself.
You will compare the algorithms on 3 different data sets:
The votes dataset from the CUI repository
A data set of your choice
A 3-of-5 Boolean function, to which you will add classification noise.
Prof. Doina PRECUP
Last modified: Wed Sep 19 17:22:09 EDT 2001