Human experts develop knowledge over time. Some of this is based on their initial training, and so may be obtianed form books or traing materuals, but much is implcit knowledge, whuch they may not even be able to fully articulate (unknown knowns). Knowledge elicitation ate pts to cpature this expert knowledge in forms that can be used for hand-crafted expert system rules, or as training data for machine learning. This may involve expert labelling of data.
Used on Chap. 14: pages 336, 338; Chap. 18: pages 422, 423, 438, 441, 448, 451