how to classify a data (URGENT)

I have 10 column of data ,5 groups (each group contains 2 colums of data)they are named as fc1,fc2,fc3,fc4,fc5.i want to use 4 groups fc2,fc3,fc4,fc5 to fit to a classifier ,a FNN ,can anyone tell how to process

[Jan]

Dear Pat, this is yout 82nd question in this forum. I expect that you have learned already, that every question is urgent for the author, and not urgent for the ones who spend time for answering. Therefor adding this term, even in uppercase, will *decrease* the motivation to post an answer.

The descritpion of your problem is hard to understand. "10 column of data" cannot be converted to a Matlab variable unequivocally. Using the names fc1, fc2, ... is not useful, as explained repeatedly: fc{1}, fc{2}, ... would allow to access the variables automatically using an index.

You do not explain the process of classification with any details.

Is FNN the "false nearest neighbors" function written by S. Mohammadi or do you mean a fuzzy neural network? What should the decision about the classification be based on?

Please, Pat, add the necessary details. Read your question twice and check, if a meaningful answer is possible.

Posting the same question with two different titles is a waste of the time of the readers:

http://www.mathworks.com/matlabcentral/answers/19763-what-will-be-the-output-of-this-classifier

32 of your answers could not be answered. It is time to consider the tips in: http://www.mathworks.com/matlabcentral/answers/6200-tutorial-how-to-ask-a-question-on-answers-and-get-a-fast-answer

[Pat]

Jan i mean fuzzy neural netwoork

i have data as

fc1 fc2 fc3 fc4 fc5

3,4 5,6 7,1 1,2 3.4

1,2 4,5 6,7 8,9 45,78

.

.

.

.

etc upto 990 rows

i have to take fc1 for testing and use other four folds fc2..fc5 to fit to a classifier a FNN(fuzzy neural network) and have to use this classifier to test fc1

Answer 1

[Walter Roberson]

So, just do it. This is no different than having been given fc2 fc3 fc4 fc5 as your data to train on, and then after training, running a test against the data in fc1.

If you are asking us how to do fuzzy neural network training, then you would have been much further ahead by reading through the MATLAB documentation and examples on that topic.

[Pat]

walter in train data i hav evalues of more than 1 is it possible to apply this to data to fuzzy neural network?wen i read about fuzzy the values were only 0 to 1?,can u tell please

[Walter Roberson]

http://www.mathworks.com/help/toolbox/fuzzy/fp715dup12.html#bq98id8

[Pat]

thank walter ,but i have been told to do classify the genes using clssifier(FNN),i need to show the classified ouput,i dont know what the output is?can u tell what it will be,the data i have fc2...fc5