From the course: Applied Machine Learning: Supervised Learning

Unlock this course with a free trial

Join today to access over 24,900 courses taught by industry experts.

Decision trees

Decision trees

- [Narrator] Another very common model used for predictive modeling is a decision tree and a decision tree, you can think of it as a bunch of if/else statements. What we're going to do is we're going to loop through each of the features and then for each column or feature, there'll be a range of values. You can sweep through the numbers of those values and determine if a value is below or above some threshold. That will separate the values into a positive or negative case. And then you can apply some metric to see how cleanly it separates the classifications. You repeat that and you basically have a bunch of if/else statements in a tree form. There are a few things that are really nice about this. They're simple, they're easy to understand. You can make them work with other kinds of data, not just numeric values, and they can work with minimal pre-processing. You don't need to standardize your data. They also have the ability to capture non-linearities. For example, linear regression…

Contents