Artificial Intelligence & Machine Learning
Label
Definition
In supervised machine learning, a label is the "answer" or the correct output for a given piece of data. It is the value you are trying to predict.
Why It Matters
Labels are the ground truth that a supervised model learns from. The quality and accuracy of the labels are critical to the performance of the model.
Contextual Example
In a dataset for a spam filter, each email is an input, and the label is either "spam" or "not spam." In a house price prediction dataset, the features of the house are the input, and the label is the actual sale price.
Common Misunderstandings
- The process of adding labels to a dataset is called "data labeling" or "annotation" and can be a very expensive and time-consuming process.
- Unsupervised learning works with data that has no labels.