Skip to content

Methodology

  • Export the Mural board as CSV.
  • Python cleans data and extracts relevant information to a new JSON structure.
  • NLP technology extracts keywords from the sticky note text strings.
    • See our ChatGPT history for how we used ChatGPT to extract keywords. This was a workaround as the simple keyword extraction tool we used was insufficient for our needs.
  • Python adds extracted keywords to the new JSON structure.
  • NLP technology matches keywords from the sticky notes to keywords from the mentorship article database.
    • See our ChatGPT history for how we generated a dummy content list with keywords and used ChatGPT to match keywords and make recommendations. This was a workaround for our limited programming skills to demonstrate how the API should work.
  • Display recommended content on mentee and mentor dashboards in the e-mentoring platform.
    • This demo uses Python and Flask to create a simple API endpoint to display the recommended content.