I've been playing around with clustering text embeddings to find similar groups of things.

Grouping similar things is useful because:

  \* For your user surveys, quickly find similar answers to group participants and find personas
  \* For your reviews, quickly group customers who have similar complaints to spot issues early
  \* For community comments, figure what the heck everyone else is saying before adding your own informed opinion

I built 3 demos as n8n workflow templates which show you how to do this:

  (1) https://n8n.io/workflows/2372-survey-insights-with-qdrant-python-and-information-extractor/
  (2) https://n8n.io/workflows/2373-customer-insights-with-qdrant-python-and-information-extractor/
  (3) https://n8n.io/workflows/2374-community-insights-using-qdrant-python-and-information-extractor/

I had a lot of fun building these use-cases and I hope you find them useful. Let me know what you think!

Jim Le Follow me on LinkedIn (https://linkedin.com/in/jimleuk) or X/Twitter (https://x.com/jimle_uk)