AIExplainer
Machine Learning Intermediate 2 min read

What is positive class?

The class or category of interest in a classification problem

In machine learning, the positive class refers to the category or label that is being targeted or predicted. For example, in a spam vs non-spam email classification problem, the positive class would be 'spam' because it's the class of interest.

Think of the positive class like a specific type of animal you're trying to identify in a zoo. Just as you would look for characteristics that distinguish a lion from other animals, a machine learning model looks for features that distinguish the positive class from the rest.

In medical diagnosis, the positive class might be 'disease present' and the model would be trained to predict the likelihood of a patient having the disease based on their symptoms and test results.

The positive class is used as the target variable in supervised learning algorithms, where the model is trained to predict the probability of an instance belonging to the positive class.

A common misconception is that the positive class is always the majority class, but in many cases, the positive class is the minority class, such as in fraud detection where the majority of transactions are legitimate.

The concept of a positive class has been around since the early days of machine learning, but its importance has grown with the increasing use of classification algorithms in various applications.

target class class of interest label of interest

Three products for different needs — explore what’s relevant to you.