From the course: Machine Learning and AI Foundations: Decision Trees with SPSS

Unlock the full course today

Join today to access over 22,600 courses taught by industry experts or purchase this course individually.

How C&RT handles nominal, ordinal, and continuous variables

How C&RT handles nominal, ordinal, and continuous variables

From the course: Machine Learning and AI Foundations: Decision Trees with SPSS

Start my 1-month free trial

How C&RT handles nominal, ordinal, and continuous variables

“

- [Narrator] Back in the 80s, C&RT had the reputation of being slow. It's very computationally intensive. Our look at continuous variables will make it clear why this was so. It's considering all possible cut points. Back in the 80s, that definitely made it slow, but computers are so much faster now that it's really no longer an issue. Continuous variables will also help us understand why C&RT always produces binary splits. Let's take a look at this example. In the case of age, again from the Titanic data set, it's going to find the youngest passenger age and place that in the node on the left and take all other passengers and put them on the right, and then it's going to consider the youngest two passengers on the left, and everybody else on the right, and then it's going to consider the youngest three, and so on and so on. So it's clearly considering many, many cut points. C&RT therefore is capable of a more precise cut point on continuous variables than Shade. Shade, as you recall,…

Contents