What about real-life classification methods?
This week we encounter some of the most important classification methods.
Although the principles are not difficult, actual, working, industrial-strength implementations often involve many nit-picky little details, and you’ll probably never completely comprehend the full complexity. What this course aims to convey is the gist of modern machine learning methods, not the gory details. What’s important is that you can use them and understand the principles behind how they work – just as you can use a car, or a computer, without being an automotive engineer or hardware/software specialist.
It’s never a good idea to blindly use methods without any understanding of what’s behind them – and this is particularly true in data mining, where it’s embarrassingly easy to fool yourself about how well your system is working.
We’ll study some pretty cool machine learning methods. First up is a venerable, tried-and-true statistical technique called “linear regression. Then we learn how this can be used in non-standard ways: classification by regression, and logistic regression. We end with the relatively recent techniques of support vector machines and ensemble learning. These are contemporary, state-of-the-art machine learning methods.
By the week’s end you will be able to use modern machine learning methods in Weka, describe – at a high level – how they work, apply them to a dataset of your choice, and interpret the output that they produce.