How to apply dimension reduction to SVM/NN?

I have a dataset where each observation has about 300 features. So essentially, I have a matrix of 1124x300 (1124 rows/observations x 300 columns/features), and another matrix of 1124x1 containing class labels (0 or 1).

I want to apply dimension reduction to it, and I found about a few algorithms like PCA, and I read about normalising my data before applying an algorithm, but I am not sure how to apply it to SVM or Neural Network and train a model.

Any help would be much appreciated, as I am pretty new to MATLAB.

[Kaushik Lakshminarasimhan]

Your question is vague. It is unclear why you want to reduce the dimensionality if your goal is classification. Perhaps you can start by trying this: https://www.mathworks.com/help/stats/svmtrain.html

Answer 1

[the cyclist]

If you have access to the Statistics and Machine Learning Toolbox, you can use the pca function.

[Vania krm]

I am trying to use PCA for me selected features + labels and then classify by SVM but :

*_ how can I apply the results of my feature selection that is including one colum of label(string) and othe colums (number)? Does pca function save the rows and just chane the colums that are my features in the matrix?

Very much thanks in advances