what is meant by classification edge??.. (Matlab2011a)

plz answer the following........

1) What is edge(ens,X,Y) ??......

what is meant by "ens.prior" , "ens.X" , similarly ens. i.e what is meant by dot(.) at there??....

Answer 1

[Ilya]

Edge can be used to assess the predictive performance of a classifier. Since you don't know how to use it, I suggest that you use classification error.

ens is an object, and when you type ens.X, you access property X of this object.

Answer 2

[Image Analyst]

1) If that edge() function is the one in the Image Processing Toolbox, it does not take a structure as the first argument. It takes a numerical image array such as a 2D uint8 variable. It also does not take any X and Y arguments.

2) Your #2 is missing.

[Oleg Komarov]

EDIT [23 May 2012, 16:14 BST - OK] merged from indipendent question

Sannia says/asks:

Actually, I was talking about the classification edge... given in statistics toolbox ->Functions -> Supervised Learning ->ensemble methods -> classification -> edge

1) the thing is, I couldn't understand the actual purpose of finding edge ??.. it is used for what actually??.. I had read regarding this in the help but unable to get the basic meaning, purpose and usage of it....

2) 2ndly.. what does it means:

ens.X,

ens.prior

i.e. what is meant by dot (.) at there??..

[Image Analyst]

The dot is what separates the structure from all the fields (members) that belong to the structure. You have a structure called ens. Now you can add fields to it that are various properties of the structure, like X, prior, etc. It's like a person could be a structure and different attributes are members of the structure, like sannia is the structure, and it has members sannia.weight, sannia.eyeColor, sannia.hairColor, sannia.race, sannia.homeAddress, sannia.phoneNumber, etc. Each one of those is a variable by itself but they are all collected under the umbrella variable called sannia.