Company or Institution

CitizenLab

Industry

Government Employees

Website

https://www.citizenlab.co

Country

Belgium

Sustainable Development Goals (SDGs)

General description of the AI solution

CitizenLab’s digital community engagement platform leverages Natural Language Processing (NLP), Machine Learning (ML), and data visualization algorithms to help local governments and organizations in two primary areas: 1) promoting participation in digital engagement projects and 2) analyzing and understanding their community’s interests in less biased and more efficient ways. This results in more inclusive, participatory, and responsive public decision-making.

As part of its analysis options, CitizenLab offers a “keyword map” which uses NLP, ML and data visualization to show the most popular keywords or concepts that are being discussed within the community. The visual representation helps community managers quickly understand what people are discussing and then group themes in order of importance. They can quickly spot trends, changes, and gaps in the project and decide whether a detailed analysis is necessary.
Using NLP, CitizenLab platforms can detect disruptive behavior and inappropriate content posted on the platform. Platform administrators are notified of this content and can moderate it accordingly to promote safe and productive participation.

Moreover, the insights map provides both highly trending and ancillary topics, helping community engagement managers easily spot side topics that emerge within the community so they can further analyze needs that might have otherwise gone unnoticed. The insights map also helps them process input intuitively and visually, instead of combing through individual posts on a tedious spreadsheet. There is also no need for community managers to have a predefined list of categories to group input, reducing sorting bias.

Lastly, leveraging a Natural Language Inference (NLI) model based on BERT and ML, the tool can understand the deeper meaning and context behind each post. It can determine whether a particular post is semantically related to a category and provide recommendations. This helps our community engagement managers efficiently process text input gathered from their community.

Github, open data repository, prototype or working demo

https://github.com/CitizenLabDotCo/citizenlab

Publications

1. CitizenLab was cited in Nesta’s report on democratic innovation and digital participation as being a tech enabler that delivers a new form of participation and as a platform designed to enable more inclusive participation.
LINK: https://media.nesta.org.uk/documents/Democratic_innovation_and_digital_participation.pdf

2. CitizenLab's content was cited in the OECD’s new report on guidelines for citizen participation.
LINK: https://www.oecd-ilibrary.org/governance/oecd-guidelines-for-citizen-participation-processes_f765caf6-en;jsessionid=Jsh0xJ8xe91SFseQJ4aSuftNsE_e7B2TZ6UsXi5k.ip-10-240-5-47

3. Empodera Impact, an SDG collective innovation ecosystem, profiled CitizenLab's COO and Co-founder, Aline Muylaert for contributing to the SDGs.
LINK: https://impact.empodera.org/impact/en/experiences/experience/citizenlab-construye-democracias-mas-fuertes-al-hacer-que-la-toma-de-decisiones-publicas-sea-mas-inclusiva-participativa-y-receptiva

4. People Powered profiled the work of CitizenLab's clients in Chile, who are running a youth engagement project with INJUV (Instituto Nacional de la Juventud, or National Youth Institute) to show how AI can be used for community development.
LINK: https://www.peoplepowered.org/news-content/chile-institute-uses-digital-platform-to-engage-youth

Needs