<img src="https://d5nxst8fruw4z.cloudfront.net/atrk.gif?account=J5kSo1IWhd105T" style="display:none" height="1" width="1" alt="">

Nexosis @ Work & Play

Forecasting Fitbit Activity

August 10, 2017 @ 1:32 PM | Technical, Use Cases


Craig wrote a tutorial using our API .NET client library and Fitbit's API to analyze his activity patterns. It is a really cool API use case if you have a FitBit.

One of the great things about APIs is that they let you, as a developer, easily combine data from lots of different sources and mash them together in a single application. The barrier to entry is often low, and you can get something up and running pretty quickly. One company that's been on the API bandwagon from pretty early on is Fitbit. As you're no doubt aware, Fitbit sells activity trackers that measure and record data such as steps taken, quality of sleep, heart rate, etc., and they expose that data through a REST API.

I've been a long-time Fitbit owner, and so have quite a bit of activity history that I've built up. I also have a decent idea of my own activity patterns, and wanted to know how well the Nexosis API can pick up on those patterns. How well can we predict my future activity level? What kinds of insights does the Nexosis API give me about my activity data? What kinds of questions does it raise? As a developer myself, I also wanted to know whether our API does a good job of accelerating how fast I can get to those insights and questions.

To those ends, I created a small web application that uses the Fitbit and Nexosis APIs to forecast activity levels for the next 30 days. Take a look!

View the tutorial »

Ready to start building machine learning applications?

Get your free API key  Talk to an expert

Craig Vermeer

Craig is one of our software engineers at Nexosis. According to his 6-year-old, he spends his days watching 'Cat vs Cucumber' videos. He also spends a lot of time writing code.