Ask Your Question

k- nearest neighbors with multivariate neighbors

asked 2013-05-05 14:04:26 -0500

kate gravatar image

I have a sample with several variables. I want to reconize leaves. So i've collected needed parameters(features), such as ratio of width and height, ratio of bounded rectangle area and a leaf area etc. All this data must be given to knn algorithm for training and test. Can anybody write a simple example?

edit retag flag offensive close merge delete

1 answer

Sort by ยป oldest newest most voted

answered 2013-05-06 16:49:44 -0500

SR gravatar image

No, you do the first step. If you have questions this is the right place. If you want somebody to code for you, you should hire a programmer.

edit flag offensive delete link more


Ok, then answer me somehow. Is it possible to perform this task with knn algorithm? Any explonations? I need an example or exlonation, how a sample with several different variables must be given to knn.

kate gravatar imagekate ( 2013-05-07 08:27:44 -0500 )edit

Theoretically yes, in practice it depends on a lot of things: Features, classes, their separability, etc...

SR gravatar imageSR ( 2013-05-07 09:00:44 -0500 )edit

all fetures are just ratios and kept in double.

kate gravatar imagekate ( 2013-05-07 13:57:44 -0500 )edit

Question Tools


Asked: 2013-05-05 14:04:26 -0500

Seen: 157 times

Last updated: May 06 '13