Build a machine learning predictor for student performance

Hi all,

My question is on how I can get Matlab to reformat raw data into a format I can use for supervised machine learning.

I have a dataset on a large number, N, of schools that lists the academic performance of students in Biology, Maths, Physics and Overall according to the school term of which there are 4 terms.

I would like to build a machine learning predictor where the performance in each of these academic subjects and the term that the school is in are input features and the school performance Overall is the target.

I would like help on Matlab commands that helps me transpose the raw dataset into a structure that I can build a supervised machine learning model from.

I have attached the raw and intended data formats and have tried using the STACK command but this is not as easy as I thought it would be. Any help would be welcomed.

Answer 1

[Saumya Goel]

I am assuming that you are having data in table format. I think the unstack function might help you resolve the above issue, you can try the below step: unstack(tablename, {'Average Grade','Median Grade','Standard Deviation','No. students','State rank'},'Term')

[Cuong Quang]

Thankyou Saumya - I will apply your suggestion and come back on whether that worked!

[Cuong Quang]

Saumya

Thank you very much for your suggestion. My final tweak to your answer is actually: unstack(tablename, {'Average Grade','Median Grade','Standard Deviation','No. students','State rank'},'Class')

This leaves the rows sorted by School No. and Term in the format I desire.