Combining Bagging and Random Subspaces to Create Better Ensembles  thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Combining Bagging and Random Subspaces to Create Better Ensembles

Published on Oct 08, 20075032 Views

Random forests are one of the best performing methods for constructing ensembles. They derive their strength from two aspects: using random subsamples of the training data (as in bagging) and random

Related categories

Chapter list

Combining Bagging and Random Subspaces to Create Better Ensembles00:00
Outline00:20
Motivation00:55
Randomization Methods for Constructing Ensembles02:01
Bagging03:23
Random Subspace Method04:04
Random Forest04:52
Combining Bagging and Random Subspaces05:52
Training Set S pt 106:20
Training Set S pt 206:24
Training Set S pt 306:34
Training Set S pt 406:55
Training Set S pt 507:05
Training Set S pt 607:11
Experiments07:18
Results pt 108:41
Results pt 210:07
Results pt 310:54
Results – Wilcoxon test11:31
Summary12:53
Further work13:25