Community Participation Analysis
Explore how public comments on local issues might be grouped into themes while keeping the limits of participation counts visible.
- Published sources
- 2
- Starter data
- 18 synthetic records
- Domain build partner
- build-community-participation
- Last reviewed
From the draft strategy
Opportunity
The draft strategy names AI for community-led social change as a potential public-service application, encouraging people to use data to change things where they live, without saying who is taking part, who is missing, or who ends up deciding.
Where the draft says it
Table 2 — potential public-service applications
Source register
Research already done
We reviewed these published sources so you do not have to start from zero. Each source shows what it covers, how it can be accessed, and why it matters.
2 sources
nidirect
OpenNorthern Ireland consultations portal
- What it covers
- The consultation platform used by several Northern Ireland departments and public bodies, listing what is open, forthcoming, and closed, with a separate portal carrying the remaining departments.
- Why it matters here
- It is where public comment on local and policy questions is actually gathered here, and the split across two portals is part of why participation is uneven in the first place.
Department for the Economy
OpenDraft Circular Economy Strategy — public consultation response report
- What it covers
- How one Northern Ireland department grouped, counted, and reported back on the responses to a public consultation.
- Why it matters here
- It shows what happens to comments after they are submitted, which is the step this example would be automating, and its vocabulary shaped how the synthetic comments are grouped.
Starter data
Starter dataset
Use 18 synthetic comments across five local topics to explore theme grouping without holding anyone's submitted words.
AI-assisted research helped identify and interpret the published sources. We then created a small, non-sensitive synthetic dataset shaped by the information those sources expose. It is for exploration—not evidence, training, or operational decisions.
- Records
- 18
- Fields
- 4
Field preview
records viewidSYN-CP-01 · SYN-CP-02 · SYN-CP-03
topicbus-service-changes · park-and-green-space · planning-and-housing
stancecritical · supportive · uncertain
textThe new timetable went up as a leaflet in the library and nowhere else. The first I knew about losing the evening bus was standing at the stop at half nine wondering where it was. If this was open for eight weeks, it did not reach our end of the road. · The extra morning service has made a real difference getting the children to school. I would only ask that it keeps running through the summer, because childcare does not stop in July. · Hard to say yet. On paper more buses is better. The ones added are at times that suit the depot rather than times anyone here is actually travelling, so I do not know whether it helps or not.
How it was prepared
AI authored fictional comments and stances to preserve disagreement and minority views found in public submissions.
Limitations
- Eighteen comments are far fewer, tidier, and more evenly spread across topics than any real set of responses.
- The five topics and four stances are this project's own choices, checked against no official framework.
- A synthetic set cannot contain the thing that matters most about participation: who did not write in, and why.
Clone and build
Domain build partner
Build with a domain-aware coding agent
Clone the repository, then ask your coding agent to load the skill that ships with it. The skill brings the opportunity, sources, starter data, known unknowns and constraints into the conversation before anything is proposed. It follows the open Agent Skills standard, so it works in Claude Code, Codex and any other agent that reads skills.
01 / Clone
git clone https://github.com/Hypership-Software/lab-notes.ai.git02 / Invoke
- Claude Code
/build-community-participation- Codex
$build-community-participation
Conditions, not a checklist
Before you build
These are not footnotes. They are conditions that any responsible prototype would need to address.
Participation counts are not public opinion
A count of comments measures who had the time, the connection, and the confidence to write, not what a community thinks.
Grouping can hide a sharp objection
Automatic grouping tends to lose the single sharp objection, and the single sharp objection is often the one worth reading.
Categories must remain contestable
People whose words are being grouped should be able to see the categories and argue with them, which is a design commitment rather than a technical one.