Embedding Matchmaking

"Words can't describe how unique your interests are... but coordinates can" - Sean Ashley, circa 2023

A flattened embedding space of names clustered based on their interests using the sentence-transformers all-MiniLM-L6-v2 model. Created for the UW Startups S23 Kickoff event with guidance from Jacky Zhao and Sean Ashley. Simha Kalimipalli later aded interactivity!

Instructions for use

1. Collect or format your data in the following format

Name	What are your interests? (or varying permutations of this question)
Alice	I love being the universal placeholder for every CS joke ever
Bob	I too love being the universal placeholder for every CS joke

2. Clone the repository
3. Install all required packages using pip or conda:

• umap-learn
• scikit-learn
• scipy
• sentence-transformers
• matplotlib
• pyvis
• pandas
• numpy
• seaborn
• branca

4. Replace attendees.csv in visualizer.ipynb with the path to your downloaded data
5. Run all cells
6. Bask in the glory of having an awesome new poster
7. Make two (!) cool interactive visualizations