What will you learn?

Hello, and welcome to More Data Mining with Weka. I’m Ian Witten, and I’m presenting the videos for this course, which is brought to you by the Computer Science Department at the University of Waikato in New Zealand. This course follows on from a previous course, Data Mining with Weka. It’s a practical course on how to use the advanced facilities of Weka for data mining. As in the previous course, we’re not going to cover programming, just the interactive interfaces to Weka. We’re going to pick up some basic principles of data mining along the way. We’re assuming that you know about a number of things that you will have learned in Data

Mining with Weka: what data mining is and why it’s useful, all the motivation, simplicity first, using the Explorer interface, popular classifier and filter algorithms, evaluating the result, interpreting the outputs, avoiding the pitfalls of training and testing sets, and the overall data mining process. We’re not going to cover any of that in this course. If you want a refresher, then you can go to YouTube and look at the WekaMOOC channel where you’ll see all the videos for the previous course. As you know, a “weka” is a bird found only in New Zealand, but from our point of view, it’s a data mining workbench – the Waikato Environment for Knowledge Analysis, which contains a lot of machine learning algorithms.

A very large number of algorithms for data mining tasks: preprocessing algorithms, feature selection, clustering, association rules – things like that. It’s a pretty comprehensive machine learning workbench. What you’re going to learn in this course is how to use the other interfaces to Weka. We already know how to use the Explorer, but we’re going to talk about the Experimenter, the Knowledge Flow Interface, and the Command Line interface. We’re going to talk about “big data” and how you deal with that in Weka. We’ll do some text mining. We’ll look at filtering using supervised and unsupervised filters. We’ll learn about discretization and sampling. We’ll learn about attribute selection. We’ll learn about classification rules, rules vs. trees, association rules, clustering, cost-sensitive evaluation and classification.

Most of all, I’m trying to get you to a point where you can use Weka on your own data, and – most importantly – understand what it is that you’re doing. Let me just finish off by saying this is where New Zealand is, at the top of the world. We think of you as being “down under”, not us as being “down under”. We’re in the top center of the world. Here in New Zealand – actually, I’ve turned this map around with North at the top, which is probably what you’re used to – you can see where the University of Waikato is, pointed to by the red arrow. That’s where I am.