Hall of Fame Project Website

WordStream++

Interactive visualization for topic evolution, based on the WordStream technique, extended with dataset management, sentiment/emotion analysis, export, accessibility palettes, and A/B period comparison.

Live Demo / Run Instructions GitLab Repository Documentation (README)

Demo Video

This is a quick demonstration of our project

What the viewer sees

  • Streamgraph by category/topic
  • Packed words per time slice (size = frequency)
  • Time axis + gridlines + zoom/tour

What we added

  • Dataset upload pipeline (optional backend)
  • Sentiment / emotion enrichment
  • A/B period comparison overlay
  • Export (PNG/SVG)
  • Color-blind palettes
  • Tour/Zoom controls

Extensions (Summary)

Generalizability

The system supports multiple datasets and an upload workflow (with column mapping and preprocessing), enabling WordStream on data beyond the original examples.

Usability

Added export, palette switching, dataset switching, and a guided animated tour to make exploration and reporting easier.

Analysis

Added sentiment/emotion enrichment via an optional NLP backend and a period comparison tool (A vs B).

Accessibility

Implemented color-blind friendly palettes (protanopia/deuteranopia/tritanopia + grayscale).

Text analysis modes

Sentiment (pos/neu/neg), Emotion (6-class), Emotion Advanced (GoEmotions/28), Topic Detection, Happiness Score

How to run

  1. Frontend only: start a local static server (because fetch() needs HTTP): 
    python -m http.server 8000
    then open http://localhost:8000.
  2. With backend (upload + NLP): run the Python API server in preprocess/, then run the frontend server.

For exact commands, dependencies, and dataset formats, see the README in the repository.

References