Skip to content
BENIBIX Learn

Regression (Machine Learning)

Also known as Regression Task, Regressor

Guessing a number from the facts you have. The number might be a price or a time.

Think of it like

Think of it like guessing how tall a pup will grow. You look at the size of its paws.

Example

Emma's model reads a flat's size, floor and street, then guesses its monthly rent. For one flat it says 1,450 euros.

Why it matters

Many business questions are how much or how many. Sales, prices, wait times and demand are all regression.

Definition

Regression is the machine learning task of guessing a number. The number can be any value on a scale, like a price, a temperature or a time. The model learns from examples where the real number is known. For a new example, it gives its best guess of the number. The name is old and has nothing to do with going backwards.
Regression is the right panel. The answer is a number on a scale, read off the line.

Common mistake

Using it when there are only a few fixed answers. If the answer is small, medium or large, you want classification.

Key takeaways

  1. 1
    A number out. Any value on a scale.
  2. 2
    Learns from known numbers. Past examples with real values.
  3. 3
    Best guess. Close, not exact.

Compare

Learn it properly