From the course: Advanced Predictive Modeling: Mastering Ensembles and Metamodeling
Unlock the full course today
Join today to access over 22,600 courses taught by industry experts or purchase this course individually.
What is Boosting and how does it work? - SPSS Tutorial
From the course: Advanced Predictive Modeling: Mastering Ensembles and Metamodeling
What is Boosting and how does it work?
- [Instructor] Rather than a bunch of definitions or mathematical formulas, I want to explain boosting via a demonstration. So first, let's just dispel a very quick myth. Almost all the discussions of boosting that you're going to see use what are called: decision stumps, decision trees with just one branch. Boosting as a general concept can be applied to any model. So I actually have a random trees model here, and what I've done is I've identified which ones I've got right and which ones I've got wrong, so that I can do a table of just the ones I've got wrong. And we're in the Titanic dataset obviously, and I have 103 records that were incorrect. So this is what I'm going to do. Boosting is all about increasing the weight of those rows that I'm getting wrong. Now, what makes the Adaboost algorithm challenging to explain is the math behind the weighting can get quite complex; so I'm going to do something a lot simpler, but it's not identical to the Adaboost. I'm going to just double…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.