AI Reporting Project at PlaceSpeak

May 27, 2025

Technologies

FlaskOpenAINltkJavaScriptPostgreSQL

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.

  1. Integrating Nltk for Advanced NLP Analysis

    • Leverage the OpenAI API to perform summary of public feedback and graphs.
  2. Designing the Data Processing Pipeline

    • Build a full-stack analytics pipeline using Python, SQLAlchemy, and SQLite to handle cached generated reports to limit cost.
  3. 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!