Global AI and Data Science

Global AI & Data Science

Train, tune and distribute models with generative AI and machine learning capabilities

 View Only
Back to discussions
Expand all | Collapse all

An interesting public IoT data set - particulate matter measurements from low budget sensors

  • 1.  An interesting public IoT data set - particulate matter measurements from low budget sensors

    Posted Tue November 06, 2018 07:53 AM
    A while ago I was pointed to a public IoT data set published by the OK Lab Stuttgart you may be interested in.
    The Open Knowledge Lab Stuttgart is member of the Open Knowledge Foundation Germany, a non-profit organization that advocates open knowledge, open data, transparency, and civil participation.

    The group dedicates its work to enable the measurement, collection and publishment of particulate matter data. Due to the fact that BAM certified measurement stations are very expensive there are very few installations. As the OK Lab Stuttgart founders felt that the small number of official measurement stations did not provide enough coverage to gain insight into the air quality of the city building instructions of how to build a sensor card from low budget sensors which measure on the one hand particulate matter values P2.5 and P10 and on the other hand temperature and humidity for less than 30 €. People all over Europe started building these cards, feeding their measurements back to the OK Lab Stuttgart, which in turn consolidates and publishes the data.

    While the crow-funded effort started out in 2015 with the goal of 300 installations in Stuttgart there are meanwhile more 5600 cards installed across Europe, the most of them of course in Germany. The data are published by the OK Lab Stuttgart here . A copy of the data is provided on IBM Cloud Object Storage ready to use and easy accessible via IBM Cloud SQL Query. The data volume is currently approximately 110 GB and growing at a rate of 3 GB per month.

    Recently the BTW 2019  announced a data science challenge  based on this data set. It will be interesting to see with what other kinds of data challenge participants will combine these data and what they will find out exploring and analyzing the particulate matter measurement data.

    If you want to work with this dataset yourself you may want to start with this  IBM Watson Studio python notebook we created to get familiar with the data.

    It would be great if you would share your ideas about interesting data sets to combine the particulate matter measurement data with, topics you find worthwhile to explore, assets (notebooks, models) you created based on this data as comments to this post.


    ------------------------------
    Ute Schuerfeld
    ------------------------------

    #GlobalAIandDataScience
    #GlobalDataScience