Binary SVM Cascade Classifier

I am trying to train a SVM with data that I have extracted from movie scenes. This data comes in .xml files and contains information about both audio and video of the scene. Now I am having problems in finding how to train a SVM in Matlab. First of all it is not clear how to get the data from the files that I have generated and get them to be correctly operated by Matlab.

Just to give more info, the classification has to be in cascade because I have got 4 movie genres. So my question is:

How can I get the data from my .xml files and train them with a SVM using Matlab?

Thank you for your time.

[Andrés Vacas]

I already got the info out of the xml files. Now I would like to know what file extension or format(.txt, .csv, .xls...) is best supported by the training commands and what command is best.

Thank you for your help

Answer 1

[Mudambi Srivatsa]

You can import data from a variety of file formats (.txt, .csv, .xls and so on) into MATLAB table format using "readtable" function.

Refer to the following link for more information:

https://www.mathworks.com/help/matlab/ref/readtable.html

Once you import the table data, you can use it in "fitcsvm" to train or cross-validate SVM as follows:

Mdl = fitcsvm(Tbl,ResponseVarName)

[Andrés Vacas]

Thank you Mudambi!! That solved my doubt.