human labelling

Terms from Artificial Intelligence: humans at the heart of algorithms

Page numbers are for draft copy at present; they will be replaced with correct numbers when final book is formatted. Chapter numbers are correct and will not change now.

Often humans are asked to label data with classifications or other information in order to guide the training of supervised learning. algorithms. This may be domain experts as part of knoweldge elicition, {crowdsourcing}} of data, for example through reCAPTCHA, or people naming their day-to-day actvities.

Defined on page 438

Used on Chap. 6: page 125; Chap. 9: page 178; Chap. 18: pages 438, 439, 451; Chap. 20: pages 494, 502

Also known as expert labelling

reCAPTCHA -- human creating training data for machine learning

Expert labelling used alongised unsupervised learning

Hybrid expert system incorporating machine learning guided by human expertise.