A pairwise constraint between two items indicating that they should be placed into diierent clusters in the enal partition. Candidate-Elimination Algorithm Mitchell's, (,) candidate-elimination algorithm performs a bidirectional search in the hypothesis space. It maintains a set, S, of most speciic hypotheses that are consistent with the training data and a set, G, of most general hypotheses consistent with the training data. ese two sets form two boundaries on the version space. See Learning as Search.
CITATION STYLE
Sammut, C., & Harries, M. (2011). Concept Drift. In Encyclopedia of Machine Learning (pp. 202–205). Springer US. https://doi.org/10.1007/978-0-387-30164-8_153
Mendeley helps you to discover research relevant for your work.