Citizen Science Project Builder

The Citizen Science Project Builder (CS PB) is a web-based tool that allows volunteers to collaborate on solving complex data classification tasks that automatic tools cannot handle. It supports projects where citizens can analyze or enrich existing data, typically large sets of images or texts, such as satellite pictures or social media posts, as well as other media formats such as videos and scanned documents.

The point of the PB is to enable people to create such projects with limited technical knowledge of crowdsourcing, and ideally little or no coding skills. The web interface is based on Crowdcrafting, a project launched in 2011 by Citizen Cyberlab, which with its underlying PyBossa software was spun out as part of SciFabric. The CS PB will be launched in early 2020. The software will be made publicly available under the ‘CitizenScienceCenter’ organisation on Github.

SDG in Progress

The SDG in Progress platform allows all innovators to document ongoing projects, or to get inspired by other people’s projects, re-use them or re-purpose them. The platform is based on Build in Progress, originally developed by MIT Media Lab. Compared to other documentation platforms (wikis, Github, etc.), SDG in Progress provides a highly visual overview of how a project was conceived, allowing for easy documentation of the sort of branching that naturally occurs in projects, where different options are explored.

The goal of SDG in Progress is to provide an open repository of SDG projects, many involving crowdsourcing and open data, where students and citizen scientists can share their results and inspire each other. The idea of SDG in Progress is to document creativity, while minimizing redundancy. Reinventing the wheel is not a problem, as long as you can improve it each time.


Citizen science powered by deliberation: As in any other project, citizen science project teams also need to self-organize, propose and discuss ideas, schedule meetings, conduct surveys, etc. decidim4cs is a digital platform for participatory citizen science. It allows citizen scientists to organize themselves democratically by making proposals, attending public meetings, making decisions through different forms of voting, and monitoring the implementations of these decisions.

decidim4cs constitutes the deliberation technology used in Crowd4SDG. It is based on decidim, a free open-source software originally created by the Barcelona City Hall as a participatory democracy platform for cities and organizations.

User Story: A citizen named Kane thinks that analyzing chemicals in drinkable water would help minimize their adverse health impacts in his city. In order to propose his methods to accomplish this objective and hear others’ ideas on it, Kane creates a project in decidim4cs. He makes two proposals: (1) Distribution of test toolkits to fellow citizens to assess the quality of water in different districts; (2) Analysis of the watersheds around the city. Kane also starts a debate on the impact of healthy water on the local economy.

decidim4cs helps Kane to promote his project on social media. Interested citizen scientists start dropping in and contributing. Some comment on the proposals—being in favor/against/neutral, some suggest new proposals, some engage in debates sharing their opinions, some just up/down vote on already made comments. After seeing the participation, Kane also decides to run a blog and conduct a survey of how tap water is used by the citizens. Upon further deliberation, one of his proposals is rejected and one accepted. For a face-to-face discussion of the ideas, he schedules a social gathering and decidim4cs notifies all project participants about the event.

User Journey: A citizen who wants to contribute to an existing citizen science project on decidim4cs follows these steps.

  1. Sign-up/sign-in to decidim4cs,
  2. Navigate to the project page,
  3. Participate in ongoing proposals, debates by commenting or voting,
  4. Create new proposals and debates,
  5. Share these activities on social media,
  6. Answer surveys if any,
  7. Monitor project’s progress,
  8. Schedule meetings

A citizen who wants to start a new project has to take the following step first.

  1. Send an email to the admin, writing his/her name and username in decidim4cs, project’s title, area, and scope.
  2. Then, decidim4cs admins will create the project and help the citizen in organizing project components (proposals, debates, etc.).

Detailed Description: Open-source code of the tool and further information can be found at GitHub.


Goodwall is a platform for students and young professionals to connect on shared interests, showcase themselves and discover learning and earning opportunities. The Goodwall community has over 1 million members across over 150 countries.

Crowd4SDG collaborates with Goodwall to promote its projects and to reach out to students.


Citizen science need covered: Crowdsourcing Citizen Science projects usually require citizens to classify items (images, pdfs, songs,…) into one of a finite set of categories. Once an image is classified by different citizens, the different votes need to be aggregated to obtain a consensus classification. Usually this is done by selecting the most voted category. Crowdnalysis allows Crowdsourcing Citizen Science projects to compute consensus that go beyond the selection of the most voted category, by computing a model of quality for each of the citizen scientist involved in the project. Those more advanced consensus result in higher quality information for the Crowdsourcing Citizen Science project.

User Story: Northwestern Albania was struck by a strong 6.4-magnitude earthquake on 26 November 2019. A Crowdsourcing Citizen Science project was created to obtain images from Twitter and use them to obtain valuable information with respect to the level of damage at different locations and different individual buildings. A part of that project required citizens to classify the images according to the level of damage that was shown in the image into one out of five different categories, namely irrelevant, no-damage, minimal-damage, moderate-damage, and severe-damage. Using crowdnalysis, the project was able to:

(1) identify effectively which were the most common mistakes when classifying an image (which are the labels which are more likely to be swapped), and 

(2) make use of that information to obtain a consensus which was proven to be higher quality than that obtained by majority voting.

As a side result the usage of crowdnalysis to obtain consensus, not only allowed the project to reach higher quality consensus from existing data, but also to plan how many experts will be required in further data collection stages  to reach an specific level of quality. 

User Journey: A Crowdsourcing Citizen Science Project willing to use Crowdnalysis will go through the following steps.

  1. Create a crowdsourcing Citizen Science Project in Pybossa (we recommend using the Citizen Science Solution Kit). 
  2. Obtain the data from the crowd. 
  3. Once the data is obtained from the crowd, export it from Pybossa into csv files.
  4. With the support or the Crowd4SDG project, use jupyter-notebook to obtain the error model from the citizens and the consensus resulting from applying those models. 
  5. Evaluate the consensus, discuss and eventually go back to 4 for changes in the data analysis, or 2 if additional data is required to increase the data quality.

Detailed Description: The tool source and more detailed description can be found on GitHub.

Image-based social sensing

Getting first-hand information about an emergency situation while it is happening is important, but is often difficult to obtain in time. The image-based social sensing toolkit allows extracting visual evidence about a situation from Twitter by searching for images posted and geolocating them.

User Story: During the COVID-19 crisis, getting evidence about the social impact of the situation is particularly difficult due to the worldwide diffusion of the virus and the variable situation in each country. It is therefore also difficult to assess the impact of containment measures. With image-based social sensing, we collected images about the usage of face masks around the world and built thematic maps for countries for which sufficient evidence was collected. With this approach, we were able to illustrate the situation in mid-August 2020, answering the question “are people wearing masks?”. The results show a good correlation with data collected by other sources through surveys, which are longer and more costly to obtain.

User Journey: With GEAR 1 Image-based social sensing you can perform the following steps:

  1. Crawl twitter with your own keywords to search for posts with images
  2. Apply selected filters (e.g., Not Suitable for Work, Photo/no Photo, outdoors)
  3. Associate locations to posts, event if tweets are not natively geolocated
  4. Posts can be evaluated within crowdsourcing initiatives using the PyBossa platform
  5. Collection of images for a location or thematic maps can be created to support interested users

Detailed Description: for further information write to