I didn't get your question properly....

If your question is regarding 'Fuzzy C-Means Clustering for Iris Data", the iris.dat file can be used as a variable in work space and exported in any required format!

If your question is different please ask ...

dataset for what?

explain your experiment .

Thank you Mr Kumar and Mohamad, for any new clustering approach, at first it should be evaluated on the benchmark data sets like Vowel, Wine, .... then if the best and mean cost becomes reasonably better than available fuzzy clustering methods then its performance is proofed. I tried to find them but in Matlab toolbox I only found iris data set, within yesterday, I could add one of datasets to the table, based on https://archive.ics.uci.edu/ml/datasets/Glass+Identification and glass_dataset.mat file in matlab. But I thought these important data sets are easily available.

Hello Farshid,

you can find many usefull datasets at: https://archive.ics.uci.edu/ml/datasets.html to benchmark your approach.

What you need is to download the data and import them in MATLAB. So they will have the format (*.mat) that you want. You just need to import the text file, for example Wine.txt in MATLAB, by clicking right on the file.

You can also use the recent dataset from my colleagues at: https://archive.ics.uci.edu/ml/datasets/banknote+authentication

Hello, Sahar,

Fantastic helping reply!,

Based on your guidance and the related tutorials, I could provide a new data set in mat format for testing the newly proposed fuzzy clustering approach,

Also thanks your colleagues,

{

filename = 'data_banknote_authentication.txt';

delimiterIn = ',';

headerlinesIn = 1;

A = importdata(filename,delimiterIn,headerlinesIn);

}

Best Regards,

Hope the best stressfulness,

Farshid