id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_qc3hpktherfxbjlacpurupzti4	Frederic Stahl	Random Prism: a noise-tolerant alternative to Random Forests	2013	22	.pdf	application/pdf	7062	617	64	noise in the training and test data, Prism algorithms can outperform decision ensemble learner using a member of the Prism family as the base classifier Random Prism ensemble learner; Section 4 evaluates Random Prism on several datasets and compares it with a standalone Prism classifier in terms of and a parallel version of the Random Prism ensemble classifier. the motivation for developing the Random Prism ensemble classifier aiming Random Prism approach based on the RF ensemble learner. then discusses the new Random Prism ensemble classifier. Fig. 2 The Random Prism architecture comprising Bagging, R-PrismTCS base classifiers and weighted majority voting. The Random Prism ensemble learner's ingredients are RDF's random feature subset selection, RF's bagging and PrismTCS as base classifier. use of bagging in the Random Prism ensemble classifier proposed here. Random Prism induces 100 base classifiers, however the runtimes are much Prism and decision tree base classifiers for each sample.	./cache/work_qc3hpktherfxbjlacpurupzti4.pdf	./txt/work_qc3hpktherfxbjlacpurupzti4.txt