AIExplainer
Machine Learning Intermediate 2 min read

What is negative class?

The class or category in a classification problem that represents the absence or opposite of the desired outcome

In machine learning, a negative class refers to the group of data points that do not exhibit the characteristic or behavior of interest. For example, in a medical diagnosis model, the negative class might represent patients who do not have a particular disease.

Think of a negative class like a 'no' or 'absent' category. Just as a 'yes' or 'present' category would represent the presence of something, a negative class represents the absence of it.

In a spam detection model, the negative class would represent emails that are not spam, while the positive class would represent emails that are spam.

Negative classes are used in supervised learning to train models to distinguish between the presence and absence of a particular characteristic or behavior. The model learns to predict the probability of an instance belonging to the negative class or the positive class.

A common misconception is that the negative class is always the 'bad' or 'undesirable' category. However, this is not always the case. The negative class simply represents the absence of the characteristic or behavior of interest.

The concept of a negative class has been around since the early days of machine learning and has been used in various forms of classification problems.

absent class non-event class null class

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