Boosting analytical potential of health tracking data using RML and Solid
Keywords: Linked Data, AI, Solid, Biohacking, Data Analysis, Decentralization, Health, Machine Learning, Mobile Health, Productivity
Supervision: Ben De Meester Anastasia Dimou
Students: max 1
More and more ways to register our daily habits become available.
- Workouts can be registered using Polar hardware, such as the Polar V800 sports watch, H10 heart rate sensors, cadans sensor, etc.
- Activity can be monitored using smartwatches, smartphones, step counters, etc.
- Nutrition can be tracked using apps like Lifesum, MyFitnessPal.
- Other health parameters such as weight, temperature, blood pressure, sleep can be tracked using, for example, Withings products.
- Productivity time tracking (e.g. https://timewarrior.net/, timely, etc.)
The main problem is the vendor lock-in.
- When we are tracking workouts using a Polar sports watch, we have to analyse our workouts on the Polar Flow platform, where all of our workout data is stored.
- Analysing our weight, temperature, blood pressure, would require us to use the Withings platform.
- Although analyzing the data individually on their respective platforms might be insightful, it would become a lot more meaningful when the data can be linked together.
Nowadays you can integrate all this data from different devices, build your own knowledge graph, store it using Solid pods and create your own analytics!
- Prior knowledge of RDF, RML, Solid are bonus or interest to learn
- Interest in Machine Learning
- Python experience is definitely useful for data analysis, visualizations & ML
- JS for Solid