random forest

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.

A random forest is a form of ensemble method for machine learning. Large numbers of decision trees are generated by choosing random subsets of features from random subsets of the training data. The indivdiual trees each give their own results and a separate layer of processing combines the results from all of the trees to give a final result. While compartively simple, random forests have proved highly effective in a wide variey of application domains.

Used on Chap. 5: page 102; Chap. 8: page 163; Chap. 9: page 183; Chap. 16: page 387; Chap. 18: page 440

Also known as random decision forest