Evaluates a classifier on held-out labelled examples
Precision and recall per class, with the confusion matrix
Can AI do this?
Not tested yet
No AI benchmark covers this knobit yet. New results are announced in the Can AI Do the Job? newsletter.
What it looks like in practice
Expected result: Precision and recall per class, with the confusion matrix
How it is assessed: Evaluation examples were not used in training, the cost of each error type is stated, and the threshold is chosen from those costs
Related knobits
Builds on
Needed for
All 6 knobits of Perform data mining.
Source: ESCO skill 'perform data mining' (ESCO v1.2.1); DO process activities P4.4, P7.6; CRISP-DM 1.0; Kohavi, Tang and Xu, Trustworthy Online Controlled Experiments (2020); the assessment criterion is this register's own and is in no reference text