Difference Between Classification and Regression

The key difference between classification and regression tree is that in classification the dependent variables are categorical and unordered while in regression the dependent variables are continuous or ordered whole values.

Classification and regression are learning techniques to create models of prediction from gathered data. Both techniques are graphically presented as classification and regression trees, or rather flowcharts with divisions of data after every step, or rather, “branch” in the tree. This process is called recursive partitioning. Fields such as Mining uses these classification and regression learning techniques. This article focuses on the Classification tree and regression tree.

CONTENTS

1. Overview and Key Difference
2. What is Classification
3. What is Regression
4. Side by Side Comparison – Classification vs Regression in Tabular Form
5. Summary

What is Classification?

Classification is a technique used to arrive at a schematic that shows the organization of data starting with a precursor variable. The dependent variables are what classify the data.

The classification tree starts with the independent variable, which branches out into two groups as determined by the existing dependent variables. It is meant to elucidate the responses in the form of categorization brought about by the dependent variables.

What is Regression

Regression is a prediction method that is based on an assumed or known numerical output value. This output value is the result of a series of recursive partitioning, with every step having one numerical value and another group of dependent variables that branch out to another pair such as this.

The regression tree starts with one or more precursor variables and terminates with one final output variable. The dependent variables are either continuous or discrete numerical variables.

What is the Difference Between Classification and Regression?

Classification vs Regression
A tree model where the target variable can take a discrete set of values.	A tree model where the target variable can take continuous values typically real numbers.
Dependent Variable
For classification tree, the dependent variables are categorical.	For regression tree, the dependent variables are numerical.
Values
Has a set amount of unordered values.	Has either discrete yet ordered values or indiscrete values.
Purpose of Construction
Purpose of constructing the regression tree is to fit a regression system to each determinant branch in a way that the expected output value comes up.	A classification tree branches out as determined by a dependent variable derived from the previous node.

Summary – Classification vs Regression

Regression and classification trees are helpful techniques to map out the process that points to a studied outcome, whether in classification or a single numerical value. The difference between the classification tree and the regression tree is their dependent variable. Classification trees have dependent variables that are categorical and unordered. Regression trees have dependent variables that are continuous values or ordered whole values.

Reference:

1.“Decision Tree Learning.” Wikipedia, Wikimedia Foundation, 13 May 2018. Available here 

Image Courtesy:

1.’Data Mining’By Arbeck – Own work, (CC BY 3.0) via Commons Wikimedia