Machine learning classification · BEGINNER LESSON
Binary label
A binary label is a classification label with exactly two possible categories, such as yes/no or present/absent. In single-label classification, it marks one sample as belonging to one of two mutually exclusive classes.
What it is
A binary label is a label used in classification when there are only two possible categories. A sample means one item being sorted, such as one patient record or one image. In single-label classification (SLC), each sample is assigned to exactly one class, so the two classes must be mutually exclusive: a sample can be in one class or the other, not both at once. [1]
[2] Single-label and multi-label classification for disease recognition ...
How it works
Binary labels are often used in binary classification, which is the simplest case of classification with two outcomes. For example, a model might assign each patient record to either “heart attack” or “not heart attack.” This decision uses the available data features, called covariates, to predict which of the two labels fits best. [1]
[2] Single-label and multi-label classification for disease recognition ...
Relationship to other classification types
Binary labels are a special case of single-label classification. Multi-label classification (MLC) is different because one sample can receive several labels at the same time, or even none. So, binary labels are about a two-choice setup, while multi-label systems allow more flexible assignments. [1]
[2] Single-label and multi-label classification for disease recognition ...
Limitations
Binary labels are useful only when the task really has two classes. If the real situation can include several conditions at once, then forcing a binary label can hide important information. The source also notes that neither single-label nor multi-label approaches are always better; which method works best depends on the problem, the number of classes, and class imbalance, meaning one class may appear much more often than another. [1]
[2] Single-label and multi-label classification for disease recognition ...
SEE THE IDEA
Binary label in classification
One item to classify, such as a record or image.
The task offers exactly two mutually exclusive choices.
The assigned output, such as yes or no.
A classifier uses features to choose one label.
Sample -> model decision -> binary label within two classes
A binary label is the output of a two-class classification task with one label per sample.
TRY IT YOURSELF
Test your understanding
Choose an answer, then check the explanation.
PUT IT TOGETHER
Match each concept to its meaning
Drag a label onto its description, or select a label and then select a description. This also works with a keyboard.
4 pairs to match
Sources and further reading
This general lesson was generated from public source material. It explains a concept; it does not interpret a private paper or provide medical advice.