(1) For general classification, data labeling identifies data such as name, social security number and driver's license. Indirect data include race, age and gender.
(2) For AI training, data labeling is the manual identification of images and categories. It is an essential component in AI development.
Data Labeling and Data Annotating
Labeling is a subset of data annotating, however, the data labeling term often applies to both disciplines. Technically, data labeling means identifying an image as a dog or flower, whereas data annotation is much broader and describes not just objects, but feelings and meanings.
All Around the World
Millions of people worldwide are involved with data labeling and data annotation. The jobs are performed in the U.S. as well as in India, Asia and South America where labor rates are lower.
Labeling and annotating are tedious jobs requiring people to sit at computers all day long. However, in some cases people are paid a much higher hourly rate than other stare-at-the-computer-all-day jobs. See
supervised learning.