Technology Forge - WEKA Tutorials

Weka is a comprehensive tool bench for machine learning and data mining. Weka is free open source software developed as part of the Weka Machine Learning Project at the University of Waikato in New Zealand, and can be downloaded along with extensive documentation from the project’s web site.

Learning about data mining can feel like drinking water from a fire hydrant. Learning how to use data mining software at the same time can feel like drowning in a sea of information. The Technology Forge Weka tutorials constitute a parallel path to the documentation provided by Weka developers that will allow new Weka users to get up and running as fast as possible. Consider these tutorials as water-fountains of information on data mining, the purpose being to provide an introduction and gradual expansion of skills in using data mining tools and the Weka data mining tool bench.

These tutorials do not replace the documentation provided by Weka developers. The tutorials are intended to encourage pursuit of data mining by providing step-by-step solutions to initially simple and increasingly complex real-world data mining problems.

Datasets

The Weka software download will include several datasets. The Technology Forge Weka tutorials use these and some additional datasets which can be downloaded here:

Assignments

Each tutorial is accompanied by an assignment designed to reinforce the learning goals of the tutorial. For students taking these tutorials as part of a course, assignments will be prepared as PowerPoint presentations which students will e-mailed to the instructor upon completion. A PowerPoint template for assignments is provided here. Each assignment has detailed instructions on what to include in the PowerPoint document submitted by the student. A solution for the assignment will be e-mailed to the student upon submission of their assignment.

Running the Tutorials

The following links provide access to the tutorials. There are several options for running the tutorials:

1. Click on the tutorial .ppt link and “Save” the .ppt to your computer. This allows you to view and print the presentation and speaker notes, but not listen to the audio narration. The speaker notes and the audio are the same.

2. Click on the tutorial .ppt link and “Run” the presentation. This allows you to play the audio, but you cannot see the speaker notes. The audio narration only runs when you are on-line.

3. Click on the .zip link and “Save” to your computer. This saves .ppt tutorial document and all the audio files to your computer. You can view the slide show with audio, and print out the slides and speaker notes.

Tutorial Developers

The primary author of the tutorials provided here is Mark Polczynski, Principal of The Technology Forge. Assistance was provided by Professor Andrzej Kochanski

and Professor Marcin Perzyk of the Faculty of Production Engineering at the Warsaw University of Technology in Warsaw, Poland. Tutorials were tested by students of Professors Kochanski and Perzyk, and students from the College of Engineering, Marquette University, Milwaukee, Wisconsin, USA.

Weka Developers

Weka's core developers are Eibe Frank, Mark Hall, and Len Trigg. Many others have made significant contributions, in particular, Remco Bouckaert, Richard Kirkby, Ashraf Kibriya, Peter Reutemann, Xin Xu, and Malcolm Ware. The primary reference for Weka is the book:

Ian H. Witten and Eibe Frank (2005) "Data Mining: Practical machine learning tools and techniques", 2nd Edition, Morgan Kaufmann, San Francisco, 2005, ISBN 0-12-088-407-0.

For problems with this web site, contact the webmaster