From the course: Data Science for Java Developers

Unlock this course with a free trial

Join today to access over 22,400 courses taught by industry experts.

Overfitting and how to avoid it

Overfitting and how to avoid it - Java Tutorial

From the course: Data Science for Java Developers

Overfitting and how to avoid it

- [Instructor] All right, so before we get into actually building our models in Java, there's one more topic that I want to cover. And that is the topic of "Overfitting," which is a very important topic because it's a trap that a lot of beginners to data science fall into. So first of all, a very general definition of overfitting is when we're trying to create a model to represent a given dataset, and we go a little too far, we make our model a little or a far too complex to the point where it's just not really very useful anymore, right? So looking at just these fake models here, let's assume that the dots on these graphs are the actual dataset. And we're trying to come up with a simplified representation, that is a model that is going to help us predict where other points are most likely to fall on that graph. So over on the left, we have a model that's really, it's over simplified, right? It gets pretty close…

Contents