What is classification and regression in machine learning?

Classification and Regression are the two main types of problems solved in Supervised Machine Learning. They sound similar, but the core difference lies in the type of output you are trying to predict. | Feature | Classification | Regression | | :---: | :---: | :---: | | Goal | To predict a category or class | To predict a continuous quantity or value | | Output Type | Discrete, fixed number of possibilities (labels) | Continuous, infinite number of possibilities (real numbers) | | Example Output | Yes/No, Spam/Not Spam, Cat/Dog/Bird, High/Medium/Low | \$45,000, 3.5 days, 98.6 degrees, 12.5 meters | ### Classification The model learns to map input data to a discrete label. * Examples: Predicting whether a customer will churn (Yes or No), identifying if a picture contains a hot dog (Hot Dog or Not Hot Dog), or determining the sentiment of a review (Positive, Neutral, or Negative). ### Regression The model learns to predict a real-valued output based on continuous input features. * Examples: Predicting the price of a house based on size and location, forecasting the temperature tomorrow, or estimating the number of days a patient will stay in the hospital.

To put it simply: If you're asking a machine learning model a "Which one?" question with a predefined list of answers, you're doing Classification. * Example: Is this email spam or not spam? (Two classes/answers). If you're asking a machine learning model a "How much?" or "What value?" question where the answer can be any number, you're doing Regression. * Example: What will the stock price be tomorrow? (A continuous number). The mathematical algorithms used for each are very different because the goal is fundamentally different!