An efficient, scalable content-based event recommendation system. It fetches event data from a Supabase PostgreSQL database, encodes the event content using Sentence Transformers, and builds a FAISS vector index to allow real-time semantic similarity search.
This notebook builds a semantic search engine that recommends similar events based on event title, description, and tags. It is built for event marketplaces, ticketing platforms, or discovery apps.
| Tool / Library | Purpose |
|---|---|
| Supabase (PostgreSQL) | Event data backend |
| psycopg2-binary | Secure DB connection |
| sentence-transformers | Text embedding generation |
| FAISS | Fast vector similarity search |
| pandas / NumPy | Data manipulation |
| python-dotenv | Environment variable management |
git clone https://github.com/aqib420/FAISS-Recommendation.git
cd FAISS-Recommendation