Box Drawings for Learning with Imbalanced Data thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Box Drawings for Learning with Imbalanced Data

Published on Oct 08, 20142226 Views

The vast majority of real world classification problems are imbalanced, meaning there are far fewer data from the class of interest (the positive class) than from other classes. We propose two machine

Related categories

Chapter list

Box Drawings for Learning with Imbalanced Data00:00
Think of…00:13
The usual way to do this...00:47
In fact, greed leads to stupid answers...01:05
Our Approaches - 101:46
Exact Boxes Algorithm - 102:36
Exact Boxes Algorithm - 203:23
Exact Boxes Algorithm - 303:31
Exact Boxes Algorithm - 403:57
Our Approaches - 204:05
Fast Boxes Algorithm - 104:08
Fast Boxes Algorithm - 204:16
Fast Boxes Dividing Space Stage04:38
Boundary Expansion Stage05:16
Summary of Fast Boxes Algorithm05:48
Experiments06:28
Baseline Algorithms06:31
Performance analysis06:41
Experimental Comparison06:59
When does Fast Boxes perform well? When data are more imbalanced. - 107:41
When does Fast Boxes perform well? When data are more imbalanced. - 207:47
Exact Boxes is frequently one of the Best Performers08:12
Summary08:40
Thank you!09:03