Machine Learning

Homework 1

Due: February 3, 2004 (midnight)

For this assignment you will familiarize yourself with the WEKA Machine Learning Software, which we will use throughout the course for testing various learning algorithms.

1. Download and install WEKA on your preferred platform. WEKA is available here.
2. Run the ConjunctiveRule classifier on each of the 7 databases supplied with WEKA and collect the output of the runs.
3. Find a database of interest to you (other than those that come with WEKA), convert it to WEKA's ARFF format, run the ConjunctiveRule classifier on it, and collect the output. See the Links page for pointers to some data repositories.
4. Prepare one table showing the following information for each of the 8 databases.
   • Number of training instances
   • Number of attributes
   • Mean absolute error for training data
   • Mean absolute error for cross-validation
5. Email to me (holder@cse.uta.edu) a ZIP file containing the following.
   • Nicely-formatted document (MSWord, PDF or PostScript) showing the raw output from each of the 8 runs, the table, and a brief description (in your own words) of the database you obtained, including where I can find it.
   • A file containing your database in ARFF format.