From the course: Data Science Foundations: Data Mining in R

Unlock the full course today

Join today to access over 22,400 courses taught by industry experts or purchase this course individually.

Classification overview

Classification overview

From the course: Data Science Foundations: Data Mining in R

Start my 1-month free trial

Classification overview

- [Instructor] We've talked about clustering, which is putting similar things in similar piles, but sometimes similar is not enough. For example, when it comes to the mail, you can group things by size or shape or postage, but that's not what you're going for. Your one job is to put things into the right box. So if you want to think about clustering versus classifying, which are probably the two most important things happening in data science, clustering is just trying to find similarities, and the data is referred to as unlabeled, because it doesn't have a correct category that you're trying to match. That also means that the answers are subjective. What that really means is, you could do things in different ways, and the usefulness, or the utility, is the outcome by which things are judged. It's also an example of what's called unsupervised learning. Unsupervised means, again, there's not a correct category here. It's…

Contents