I am doing a project where I would like to predict some characteristics from large data set and I am expecting that I would should use some Machine Learning techniques, but not sure how to proceed. I have a background in Econometrics, but I have not really done projects like this before.

I have a large data set consisting of hour-based electricity consumption for around 200,000 households over a couple of years. For about 1,000 of these households, I know that they have a specific heating device that shows a specific pattern in the consumption (i.e. large variation from hour to hour, and through the seasons etc.). I would like to predict which of the remaining household’s consumption shows similar behavior with at certain probability. In other words, I would like to know which households uses the specific heating device.

Which techniques are recommended for this type of analysis? I know that there probably isn’t any straight forward solution, but I would really like to know where I should focus my research.

  • 1
    $\begingroup$ Create features that represent your hypothesis (max-min, daily variance, yearly variance...) and use simple ML algorithms (even decision tree) to test it $\endgroup$ – Guy Jun 29 '13 at 19:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.