Hi list,
I'm trying to use OneClassSVM to classify if a short text is a (restaurant)
review. The idea is
to build a OneClassSVM to check if the text is outlier, (so -1 -> outlier and 1
-> review).
But i got very strange result when predict on the training data: it predict
almost all the training data
as -1. but i expected only ~10 (since nu=0.1) predicted as -1.
You can find the training data (restaurant reviews in Dutch) on
https://gist.github.com/tpeng/3175cf497577fc405164 and my simple
program on https://gist.github.com/tpeng/ae955b9047a0ed589df0
Am i doing very stupid thing in setting up the parameter or choose the kernel?
Thanks,
--Terry
------------------------------------------------------------------------------
Get your SQL database under version control now!
Version control is standard for application code, but databases havent
caught up. So what steps can you take to put your SQL databases under
version control? Why should you start doing it? Read more to find out.
http://pubads.g.doubleclick.net/gampad/clk?id=49501711&iu=/4140/ostg.clktrk
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
