1

This figure from (1) shows the relation between LR and linear CRF: linear CRF is pretty much the generalization of logistic regression to sequences:

enter image description here

Is there any established generalization of random forest or SVM to sequences?

(1) Sutton, Charles, and Andrew McCallum. "An introduction to conditional random fields." Machine Learning 4.4 (2011): 267-373.

Franck Dernoncourt
  • 42,093
  • 30
  • 155
  • 271
  • RF and SVMs are very different, there is no generalization going from one to another. RF operates an ensemble of trees while SVM operates support vectors while maximizing the margin. – Vladislavs Dovgalecs Sep 15 '15 at 23:37
  • @xeon shall I split it into 2 different questions, one about SVM, one about RF? – Franck Dernoncourt Sep 16 '15 at 00:24
  • Yes and no. The difference with CRF and Logistic Regression that they both can be cast in a common framework and *only* then differences/similarities are pointed out. To my knowledge RF and SVM cannot be easily cast into a same framework. – Vladislavs Dovgalecs Sep 16 '15 at 01:16

0 Answers0