I've taken several months worth of data on health(water intake, activity, steps, sleep quality, heart rate, etc.) and I would like to use Mathematica to find correlations within the data. I would like to find how all these things affect sleep quality, and then how those combined with that sleep quality affect resting heart rate the next day. I have a fair amount of general experience in Mathematica, but none in the machine learning area. Any suggestions on how to get started?

Also if possible I would like to look at overall trends. Essentially I would like it to look at not just one day back, but multiple days back, with some sort of weighting function that makes more recent days more heavily weighted than days further back. That way I can find not just what will change my resting heart rate for tomorrow, but find what lifestyle choices can make my resting heart rate the lowest in the long run.

  • 1
    $\begingroup$ You may not need machine learning to find those correlations - checkout the documentation for Correlation and friends. $\endgroup$
    – Carl Lange
    Oct 10, 2018 at 21:26
  • $\begingroup$ I've been using that to find which variables are likely to be most useful in my models. I've just found the Predict function and it's 2orking out pretty well so far. $\endgroup$
    – Daniel
    Oct 10, 2018 at 22:00


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.