Crowdsourced data collection for Crowd4SDG


Climate change will affect all of us, but not all in the same way. It will likely lead to inequalities within and across nations and between current and future generations. Solutions that help increase the preparedness for climate change of communities, especially those most vulnerable to extreme climate events (floods, droughts, fires etc) are needed.

In recent years, technological improvements have changed the paradigm of data collection. Smartphones, wearable devices, cars, satellites, etc. can potentially collect, process, and analyze amounts of data never seen before but most of all, they can capture data, where sensor infrastructures have not yet been installed, thanks to direct participation of involved citizens. As a result, every user can potentially contribute to the data collection and support the analysis of complex phenomena, such as environmental modelling and predictions by leveraging data coming from different sources and from many different perspectives.

Nevertheless, questions of validation of data provenance, and safe distribution, analysis, and sharing of data and results are crucial to ensure that the collected data is relevant, trustable, usable. For instance, The data is produced and owned by individuals (i.e., the data from wearables devices) and requires solid, scalable trustable edge computing architectures to be collected and organized.

Several crowdsourcing technologies have been developed developed, for example by the partners of Crowd4SDG European project that CERN is participating in (, for gathering and classifying data from publicly available sources that can affect local, national and regional policies, by providing evidence of problematic situations that communities are facing. 

To tackle this challenge, you will need to:

1)  Choose a suitable crowdsource data collection and distribution tool and motivate your choice by describing the criteria and features on which the choice was made

2)  Provide a high-level but comprehensive blueprint of a crowdsourcing computing architecture showing the functional components, example dataflows, how the aspects of access and authorisation, validation of provenance, protection of confidentiality would need to be taken into account

3)  Identify a local community you can engage with and involve in the project around a simple use case, where the crowdsource data platform you have designed would enable direct participation of the community members.

4)  Make a proof-of-concept implementation based on at least a client component to collect data and pre-process data as necessary, a server component to receive, aggregate, analyse and visualise data

5)  Describe the potential impact that the solution would have on the community

Note: Teams taking part in this challenge are invited to create links with teams taking part in the Data Visualisation challenges. The creation of synergies across projects will be considered a plus in the final assessment.