How does SVM ONE_CLASS work?

asked 2015-04-14 09:12:14 -0600

thdrksdfthmn gravatar image

I am interesting in the performances of SVM with one class. Has anyone done something like this? I have read this question but it seems that it's just me who commented it. Any info would be helpful. Thanks

edit retag flag offensive close merge delete

Comments

did you have any look in this example here. It seems that there is a working example that you might be able to use.

theodore gravatar imagetheodore ( 2015-04-14 09:50:12 -0600 )edit

Hmm... Is it going to work as a classifier of licence plates, if I have a detector, that may detect the audi circles sign or part of the wheels or the "grill" in front of the car?

thdrksdfthmn gravatar imagethdrksdfthmn ( 2015-04-14 09:57:17 -0600 )edit

I have tested it, but it seems that it is not as good as the one that has 2 classes (positives and negatives). Maybe if I add much more samples...

thdrksdfthmn gravatar imagethdrksdfthmn ( 2015-04-16 08:40:35 -0600 )edit
add a comment