Knowledge Graph

The knowledge graph connects entities through typed relationships, enabling multi-hop reasoning and contextual queries.

Structure

The graph consists of:

Nodes — Entities (people, places, things)
Edges — Relationships between entities (works_at, lives_in, prefers, etc.)
Communities — Clusters of closely related entities

Graph reasoning

When use_graph_reasoning=True is set on a recall query, Memorer traverses the knowledge graph to answer questions that require connecting multiple pieces of information:


# Multi-hop query: "What tools does the engineer at Acme use?"
# Graph path: Acme Corp -> works_at -> Alice -> prefers -> Python, VS Code
results = user.recall(
    "What tools does the engineer at Acme use?",
    use_graph_reasoning=True,
)

Graph visualization

You can get the full graph data for visualization:


viz = user.knowledge.graph(limit=200)
print(f"Nodes: {len(viz.nodes)}")
print(f"Edges: {len(viz.edges)}")
print(f"Communities: {len(viz.communities)}")
 
if viz.stats:
    print(f"Total entities: {viz.stats.total_entities}")
    print(f"Total relationships: {viz.stats.total_relationships}")

Memory stats


stats = user.memories.stats()
print(f"Total memories: {stats.total_memories}")
print(f"Direct: {stats.direct_memories}")
print(f"Derived: {stats.derived_memories}")
print(f"Relationships: {stats.relationships}")

Ingesting structured data

Beyond remember(), you can ingest larger bodies of text:


user.knowledge.ingest("Long document or conversation transcript...")

This processes the text, extracts all entities and relationships, and adds them to the graph.