7

I need a dataset that can be used to build imputation models upon and that is also a "common-sense" data such as countries econometrics, Forbes2000 etc. That is, I am not interested either in customer survey datasets or sensor dataset. I have already tried to compile countries econometrics dataset which include fields such as gdp, gdp per capita, population, internet penetration rate, poverty etc. to impute missing poverty rates for example.

Unfortunately, all of the different techniques I have tried failed utterly and produced no meaningful models. I have also tried Forbes2000 datasets to predict market value of a company based on other available information on the aforementioned dataset. Although somewhat better than the econometrics dataset, this also failed with no reproducible models with a high predictive power.

Now I am stuck, since I can not find any other dataset that has "common-sense" data (for a lack of a better word) and also machine-learnable. If you could advise any hints, it would be appreciated.

1 Answer 1

0

If you are developing an ML model, the cool way to find relevant data is to launch a data search in upgini library.

It searches for relevant features for the ML model through data about 700+ mln phone numbers, 400+ mln emails, 239 countries, and up to 41 years of history.

2
  • Looks like exactly what I was searching for. Gotta check it out in detail.
    – Ahmadov
    Jul 18, 2022 at 13:22
  • Hey @aftermidnight, your answer sounds a lot like a marketing message. Please clearly state your affiliation with upgini in your answer, otherwise your posts will probably get deleted as spam. Jul 21, 2022 at 20:37

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.