Interesting data sets

[PlopSofa]

DC has a chance to work on a data set for a project for A-level.

DC is completely stuck what to choose to use. DC has been sent to Kaggle among others as potential places to download a data set.

DC wants to find a data set that would provide something truly helpful. Would anyone have any suggestions of unmined data sets that would be genuinely useful and bring to light some new information?

I know it's a bit of random one, this thread, but DC doesn't want to work on something that's already obviously known.

I've looked around Kaggle myself but I haven't found anything yet but that's probably because I'm not very clued up on these things myself either!

Could any data scientists or researchers or similar advise?

watching my kid I often wondered if there is a crossover between thumb suckers that go one to be future smokers or vapers.

Or is that just me? 😂😂

I advise students on selecting datasets for projects (at university), and we send students to Kaggle too. I often suggest using something either local or of interest (or both). Is your DC interested in sports, music, film, health, child development, business needs, politics, fashion, science, cybersecurity, history, something else? Trawl Kaggle for their field of interest. Alternatively, local data from local authorities can be less mined (google "XX council open data", or similar). I work with environmental datasets (which are often small and messy but interesting) including those shared by DEFRA / Environment Agency, eg "what's the quality of water in your local river?" The water companies have been releasing data onto a new platform this summer which I've not looked at yet.

If you say what their interests are, I might have some ideas (or not, depending).

@PlopSofa when my son was doing a year 12 project he used a dataset from the Office of National Statistics: https://www.ons.gov.uk/

Not sure I would describe their stats as "un-mined" though. Is that definitely a requirement? There are always new questions that can be asked of big datasets.

Thank you @DataQuestion and @bizzymizzy for the great suggestions. Lots to think about from your posts 🙏

Has he looked at the Data available from the UK Data Service?

ukdataservice.ac.uk/