Editorial

The first paper in this issue of the Journal of Classification is by Tri Le and Bertrand Clarke, where they show that several popular methods used for creating classifiers from ensembles (e.g., bagging, random forests, boosting, etc.) are special cases of a Bayes classifier. The implications for this research are wide-ranging, from the need to choose an ensemble classifier whenever possible to the possibility for cross-disciplinary application. Specifically, it appears that some of the theorems discussed within would have direct bearing on the “wisdom of the crowds” field ofresearch (see Surowiecki, 2004). Related to the Journal of Classification, this is the third paper in three issues that deals with classification and clustering from a Bayesian perspective (the others being Ligtvoet, 2017, and Payne and Mallick, 2018)—a welcome expansion and diversification.