Hi Lyes,
You can use “patternnet” to create feed forward neural network that can be trained using the “train” function to classify inputs according to target classes. Both these functions work for multi-class classification. You can refer to the following example of wine classification which has 3 classes - https://www.mathworks.com/help/deeplearning/ug/wine-classification.html
You may refer to the following page to understand about data preprocessing and how to do so in MATLAB - https://www.mathworks.com/help/matlab/preprocessing-data.html
Please refer to the following pages for more information regarding “patternnet” and “train” functions:
