Newbie Interested In Quantified Self: Also Looking for Personal Data Set Examples

Quantified Self QS newcomers
1

Hi everyone - I’m currently a college student majoring in engineering. And, recently a couple friends and I started realizing how much data we passively collect using Apps (such as SleepCycle, RunKeeper, Spotify, MyFitnessPal, Wunderlist etc.) and how useful it could be to link all this data to manual inputs such as happiness, stress, and productivity for the day. Well it turns out we weren’t the first to think of this because this whole QS community already exists :smiley: .

Anyways we wanted to make a free web app at my schools hackathon next weekend. This web app would be a tool for users to find predictions and correlations among their personal data. (Useful for people who aren’t stat or compsci gurus).

The problem is we only have a few weeks of data that we’ve collected, so I was wondering if anyone here would be willing to share their personal data (or a subset of their personal data) so that we could have a larger data set to test our tool on. (All data would be kept totally anonymous and we would delete it after the hackathon.) You’d also get our analysis results of your data.

Also even if you don’t have data to share advice is also welcome!

Lastly, would this tool be useful? Because I couldn’t find a similar tool when I looked around. There are a lot of apps that graph personal data but couldn’t find any that would do stat analysis on personal data from multiple sources.

Thanks
Josh McMenemy

2

For a hackathon, I’d suggest using Human API.

What do you mean by “do stat analysis”?

Here are a few services that aggregate health and fitness data so you can do something useful with it: https://zenobase.uservoice.com/knowledgebase/articles/360890-how-does-zenobase-compare-to-other-services

3

Thanks for the Human API tip that looks like it could be useful!

By ‘stat analysis’ I mean finding correlations and predictors to a variable of interest, such as happiness or productivity, from personal data (and taking into account the influence of other variables so your not just running correlations between two variables in isolation). Or comparing subset means, such as comparing total avg sleep quality to avg sleep quality when you also exercised that day.