Hi all, I have some data sequences, which are simple one dimensional data. When some patterns occur in the sequence, for example, the curve drop down with a certain slope and then quickly comes up, some event is usually associated with it. I would like to do a binary classification, where 1s will denote the happening of such events. My question is, what classifier should I use for solving the problem?

I'd like to post a figure with one curve here. However, it seems that the system only allows using link from other websites.

asked Feb 26 '14 at 11:33

Patrick%20Yan's gravatar image

Patrick Yan
16112

I can see that HMM may be applied here. However, will that be an over kill?

Will kNN be enough for the application?

(Feb 26 '14 at 15:57) Patrick Yan
Be the first one to answer this question!
toggle preview

powered by OSQA

User submitted content is under Creative Commons: Attribution - Share Alike; Other things copyright (C) 2010, MetaOptimize LLC.