Knowledge-graph explorer

Graph Laboratory is the knowledge-graph explorer of the chaipat.cc research network, served at graph.chaipat.cc. It is a research workspace for exploring large system-of-systems datasets, such as automated driving, aerospace and healthcare architectures, as force-directed graphs.

Design

The README describes a "paper-first" interface built to stay readable when a graph has hundreds of sub-modules and feedback loops:

• Radial discovery: exploration starts at a root and reveals the structure one level at a time, instead of drawing everything at once.
• Node registry: a searchable index of every node, to jump straight to any part of the architecture.
• Focus: the camera centres on the active module, so the view stays steady while a branch expands.
• Perspective gallery: thumbnails of saved viewpoints, for switching between views of the same system quickly.
• Toolbox: one sidebar holding resets, the gallery and the registry.

Built with

Next.js and React in TypeScript, with PostgreSQL and Redis behind it. The automated-driving domain connects it to his ADS safety-testing ontology from JAIST.

See also

• Ontology and knowledge graphs
• Web applications