Home › Glossary › Evaluation › Decision threshold

Beginner · Evaluation

Decision threshold

Visual diagram · (in preparation) · Math · (in preparation) · Worked example · 3 difficulty levels.

TL;DR. A decision threshold is a value used in classification to convert probabilities into discrete class labels.

Technical Definition

A decision threshold is a value used in classification to convert probabilities into discrete class labels.

How it works

A decision threshold, also known as a classification threshold, is a value used to convert the raw output of a classifier (often a probability) into a binary class prediction. If the classifier's output is above the threshold, it's assigned to one class; otherwise, it's assigned to the other. Adjusting this threshold can impact the trade-off between precision and recall.

Related Concepts

  • Binary classification — A machine learning task that categorizes input into one of two mutually exclusive classes, like 'spam' or 'not spam'.
  • Classification threshold — A cutoff value used in binary classification to convert a model's raw output score into a final class prediction.

Further Reading

  • Google ML Glossary