The process of adding answers or categories to training data.
Data labeling is like putting stickers on school lunch trays. Pizza gets a pizza sticker, not a mystery slime sticker.
It powers supervised training, tests, and fine-tuning. Bad labels teach the model the wrong lesson.
Supervised Learning
Data Labeling gives Supervised Learning the answer key for each example.
Human-in-the-loop
Human-in-the-loop helps create, check, and fix labels.
Data Augmentation
Data Augmentation can make more training data from labeled examples.
ImageNet
ImageNet became powerful because people labeled a huge number of images.