NN input creation for Document Classification MATLAB

Greetings,

I've been doing my first big work in Matlab and now I have problem that I don't know how to solve :s

Goal is use NN to classify input document made of strings. After treat all data what I have are several binary (0/1) matrix that represent 70 news of each of 10 types that I have.

So basically I have 10 different types of news and for each I have 70 examples with size 316 resulting in matrix that I have now: 316x1*70 x 10 (22120x10). As targets I have matrix 10x10, since I have 10 different categories. How must be my input matrix for it to work? Since if I use one I wrote above It gives me memory problems

Regards

Answer 1

[Greg Heath]

The input matrix has N examples of I-dimensional input vectors

[ I N ] = size(input)

The target matrix has the N corresponding examples of O-dimensional target vectors.

[ O N ] = size(target)

For classification, the target columns are columns of the O-dimensional unit matrix.

Hope this helps.

Thank you for formally accepting my answer.

Greg

[Nika]

Thanks, it took me some time to figure it out since i'm new to work with matlab, especially in those dimensions but i figured it out. But thanks for answer!

Regards