unsupervised learning

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.

In unsupervised learning the training set is unlabelled, that is it has no markers to indicate a desired grouping or value. Unsupervised learning algorithms look for commonalities to group or infer other properties of the data. This may be used on its own, or be used as the first stage before human labelling or for a subsequent supervised learning method.

Used in Chap. 5: page 68; Chap. 6: pages 90, 93; Chap. 8: pages 113, 117, 118, 124; Chap. 9: pages 127, 128, 129, 134, 136, 139; Chap. 10: page 151; Chap. 12: page 197; Chap. 14: page 229; Chap. 16: page 260; Chap. 18: pages 302, 303, 306

Also known as unsupervised, unsupervised algorithm, unsupervised machine learning, unsupervised technique, unsupervised training