confidence rating

Terms from Artificial Intelligence: humans at the heart of algorithms

The glossary is being gradually proof checked, but currently has many typos and misspellings.

During knowledge elicitation or expert labelling human experts may be asked to assess their level of confidence in a rule, argument or classification. This confidence rating may be a numerical scale, such as 0-10, a percentage, or more qualitatively in a Likert scale.

Used in Chap. 18: page 283