Knowledge graph
A knowledge graph is a structured representation of information where entities (people, papers, concepts, places) are stored as nodes and the relationships between them as labelled edges. Search engines, recommendation systems, and research tools use knowledge graphs to traverse meaning rather than match keywords.
In research workflows, a knowledge graph lets you ask questions like "which papers cite this method?" or "which concepts connect these two authors?", questions that flat note collections cannot answer without manual cross-referencing.
Atlas builds a personal knowledge graph from your notes, sources, and chats. Every citation, link, and tag becomes an edge; every paper, concept, and note becomes a node. The graph is the substrate that makes maps, search, and synthesis traceable to source material.
Compare knowledge graphs against tag systems (which only support flat one-to-many membership) and folder hierarchies (which only support strict tree containment). Graphs allow many-to-many, typed, and bidirectional relationships, which more closely mirror how research ideas actually relate.