Skip to content

Latest commit

 

History

History
33 lines (20 loc) · 876 Bytes

Readme.md

File metadata and controls

33 lines (20 loc) · 876 Bytes

Twitter sentiment on GCP using dataflow

This is a project made for fun to see how hard it would be to analyze tweets in real time for a collection of hashtags. I made this to monitor crypto tweets! (but can obviously be used for anything)

Result

This is just a simple dashboard to illustrate results.

Demo

Architecture

This project uses:

  • App engine to ingest tweets
  • Dataflow to process
  • Pub/Sub to connect the dots
  • Bigquery for storage/warehousing
  • NLP API for sentiment analysis
  • Data studio for visual analysis

architecture

Requirements

License

MIT