Decision Tree Classifier — source pixabay

You might have come across the term “CART” — it stands for Classification And Regression Trees. Classification Tree’s help us classify our data into different categories like — “Male or Female”, “Apple or Orange”, “Yes or No” etc. Where as Regression Tree’s help you predict outcomes like “Salary of a person”, “Temperature”, “Price of a house” etc.

#100DaysOfMLCode #100ProjectsInML

We have covered Regression Decision Trees in our Project 5. Let’s understand Classification Tree’s in this project.

Below is a scatter plot which represents our dataset. Since we are talking about classification, these points are classified into 2 categories — “Green” and “Red”.

image source: A-Z Machine Learning Udemy

When we run the decision tree algorithm, it will split our data into different segments. Each segment is called a leaf. The final leaves are called “Terminal leaves”. The splits are done in such a way as to maximize the number of points in a certain category. Each split tries to minimize entropy. It is a very complex concept — you can read up on it to understand the math behind it.

--

--

Omair Aasim

Passionate about building products — An advocate of AI, a software engineer by profession — an entrepreneur at heart and a sports enthusiast.