Changes at "Lightning talks - Round 2: Artificial Intelligence"

Visualizing meaning and participation at scale in Decidim data with cosmos.gl

1/ AI on participation platforms: A whirlwind tour

Speaker: Matt Stempeck, Civic Tech Field Guide

In this lightning talk, I'll give a whirlwind tour of how digital participation platforms are deploying AI in the wild with examples from around the world.

Based on my research for the Civic Tech Field Guide / People Powered, I'll cover all the use cases, spotlighting particularly creative features, and the tech behind them.

Language of the session: English

2/From Chaos to Insight: automatic proposal classification and recommendation on Brasil Participativo with BERTopic

Speaker: Bruna Pinos, LabLivre

Brasil Participativo, Brazil’s national participation platform based on Decidim, collected over 36,000 civic proposals in a few months. Manually organizing this input is unfeasible due to scale, expert knowledge demands, and alignment with official taxonomies.

To address this, we implemented a BERTopic-based pipeline to classify proposals using Brazil’s controlled vocabulary (VCGE), enhanced by expert seed words and LLM validation. The resulting clusters align with institutional categories, enabling automated, detailed reports for decision-makers. This structure also supports cross-process analysis, reveals shared subtopics, and lays the groundwork for future recommendation systems that assist voters and proposal authors. It helps governments deliver broader, theme-based feedback, enhancing transparency and responsiveness.

This lighting talk presents the technical pipeline, results, and how semantic clustering can enrich Decidim-based platforms.

Language of the session: English

3/Visualizing meaning and participation at scale in Decidim data with cosmos.gl

Speaker: Diego Arredondo, doia.studio

Tracing participation and text contributions at scale within any Decidim instance reveals meaningful emergent patterns. Visualizing proposal and comment data as contribution networks and semantic clusters can enable our understanding of how participation and priorities evolve over time.

By combining powerful interactive visualization (cosmos.gl) and topic modeling (BERTopic) techniques, using open multilingual LLMs, we can reconstruct a wide, dynamic and context-rich view of any Decidim instance, showing how participation is articulated and fluctuates across topics and processes.

This talk aims to showcase large scale visualization of Decidim data in different languages and contexts, as well as to advance the community conversation on how to incorporate sense-making and network analysis tools as reflective interfaces for a shared understanding of collective priorities.

Demos:

Tracing topics within Decidim Barcelona ( https://decidim-topics.vercel.app/ )

Networks VS embeddings visualization Decidim Barcelona ( https://www.linkedin.com/feed/update/urn:li:activity:7298447901549096962/ )

Language of the session: English