In this tutorial, you'll visualize facts about Space Exploration and Astrophysics as a semantic knowledge graph using Google Cloud SQL (PostgreSQL), pgvector, and pyvis.

The steps performed include:

• Configure a Google Cloud SQL (PostgreSQL) instance.
• Enable the pgvector extension for vector similarity search.
• Generate semantic embeddings for text data using sentence-transformers.
• Store and query vector data in PostgreSQL.
• Visualize the relationships between data points using pyvis.

This tutorial uses billable components of Google Cloud:

• Cloud SQL: Costs are associated with the instance size and storage used.

Learn about Cloud SQL pricing and use the Pricing Calculator to generate a cost estimate based on your projected usage.

• Vector Search: Uses pgvector to find semantically similar facts based on embeddings generated by sentence-transformers.
• Interactive Visualization: Renders a dynamic, physics-based graph using pyvis where nodes are facts and edges represent semantic similarity.
• Secure: Configured to use Google Colab's userdata for secure credential management.

Before running the notebook, ensure you have:

1. Google Cloud Project: A GCP project with billing enabled.
2. Cloud SQL Instance: A PostgreSQL instance (version 15+) with pgvector supported.
3. Database Credentials:
   • Connection Name (e.g., project:region:instance)
   • Database User
   • Database Password
   • Database Name