Difference between classification and regression?

In machine learning interviews, classification vs regression is one of the most frequently asked fundamental questions. The key difference lies in the type of output they predict.

1. Classification

Definition:
Classification is a supervised learning technique where the model predicts a category or class label.

Output Type:

Discrete values (labels/classes)
Example: Yes/No, Spam/Not Spam, Cat/Dog

Examples:

Email → Spam or Not Spam
Medical diagnosis → Disease present or not
Image recognition → Cat, Dog, Bird

Algorithms used:

Logistic Regression
Decision Trees
Random Forest
SVM (Support Vector Machine)
Neural Networks

Evaluation Metrics:

Accuracy
Precision, Recall, F1-score
ROC-AUC

2. Regression

Definition:
Regression is a supervised learning technique where the model predicts a continuous numerical value.

Output Type:

Continuous values
Example: Price, Temperature, Salary

Examples:

House price prediction → ₹50 lakhs
Weather forecasting → 32.5°C
Salary prediction → $80,000/year

Algorithms used:

Linear Regression
Polynomial Regression
Decision Tree Regressor
Random Forest Regressor
Gradient Boosting Regressor

Evaluation Metrics:

Mean Absolute Error (MAE)
Mean Squared Error (MSE)
Root Mean Squared Error (RMSE)
R² Score

Key Differences (Interview Table)

Feature	Classification	Regression
Output Type	Categorical (labels)	Continuous (numbers)
Goal	Assign class	Predict value
Example	Spam detection	House price prediction
Algorithms	Logistic Regression, SVM	Linear Regression, RF Regressor
Evaluation	Accuracy, F1-score	RMSE, MAE, R²

Interview Tip (Very Important)

A common trick question:

“Is Logistic Regression used for classification or regression?”

✔ Answer: Classification
Even though the name says “regression,” it is used for binary classification problems because it outputs probabilities using the sigmoid function.

Simple Memory Trick

Classification = Class labels (C for Category)
Regression = Real numbers (R for Range/Real value)