From the course: Machine Learning and AI Foundations: Classification Modeling
Join today to access over 22,600 courses taught by industry experts or purchase this course individually.
Ensembles
From the course: Machine Learning and AI Foundations: Classification Modeling
Ensembles
- [Instructor] There is a frequent claim that's made that not just some but all of the kaggle.com winners on the kaggle.com data mining modeling competition website are ensembles. So let's take a moment to talk about ensembles. Now, keep in mind when you take a bunch of models and combine them, you've just turned your model into something more complex and almost certainly into something opaque. So I don't think you want to assume on real world projects that you'll always use an ensemble, but there's a reason why they win these competitions, they're very powerful. So in its most basic form, a so-called heterogeneous ensemble would just be a combination of three models. You could have a logistic regression, a neural network and a support vector machine, or any other combination of the algorithms that we've talked about, the notion of it being heterogeneous is that we're not making an assumption here that all three of these models are trees. They are a mix. And we can simply calculate an…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.