In April 2025, I had the opportunity to collaborate with PlaceSpeak and a team of BCIT Computer Systems Technology students as part of the Industry Sponsored Student Projects (ISSP) program.
Check out the full presentation here.
PlaceSpeak is a privacy-protected civic network that connects individuals and organizations for meaningful, location-based engagement. Designed to combat bots and trolls, it ensures legitimate feedback to inform policy decisions.
Our goal
Our goal was to build an AI-powered analytics system that transforms raw public consultation feedback into meaningful, actionable insights.
-
Integrating Nltk for Advanced NLP Analysis
- Leverage the OpenAI API to perform summary of public feedback and graphs.
-
Designing the Data Processing Pipeline
- Build a full-stack analytics pipeline using Python, SQLAlchemy, and SQLite to handle cached generated reports to limit cost.
-
Enabling Dynamic Report Generation
-
Sentiment distribution charts (pie & bar graphs)
-
Keyword clouds to highlight frequent topics
-
Top phrases analysis using N-grams for contextual insights
-
Implement PDF report generation using jsPDF, allowing users to download insights for further analysis.
-
The Product
By the end of the project, we delivered a fully functional analytics dashboard that fulfilled all of the goals above and allowed PlaceSpeak to:
- Analyze large volumes of public feedback from their network using AI
- Visualize trends through interactive charts and word clouds
- Generate downloadable reports for users and clients
Conclusion
Working on the PlaceSpeak-BCIT ISSP project was an incredible learning experience. It reinforced the importance of AI-driven analytics in public engagement and allowed me to contribute to a solution that helps organizations make data-driven decisions.
I’m excited to continue exploring AI, NLP, and data visualization in future projects!