Advanced Structures

Knowledge Graphs

Instructor can be used to extract structured knowledge graphs from text. A knowledge graph represents entities and their relationships, making complex information easier to understand and visualize.

import instructor
from openai import OpenAI
from pydantic import BaseModel, Field

Initialize the client with instructor

client = instructor.from_openai(OpenAI())

Define the node structure

class Node(BaseModel):
    id: int
    label: str
    color: str

Define the edge structure

class Edge(BaseModel):
    source: int
    target: int
    label: str
    color: str = "black"

Define the knowledge graph structure

class KnowledgeGraph(BaseModel):
    nodes: list[Node] = Field(..., default_factory=list)
    edges: list[Edge] = Field(..., default_factory=list)

Extract a knowledge graph from text

def generate_knowledge_graph(input_text: str) -> KnowledgeGraph:
    return client.chat.completions.create(
        model="gpt-3.5-turbo",
        messages=[
            {
                "role": "user",
                "content": f"Create a detailed knowledge graph for: {input_text}"
            }
        ],
        response_model=KnowledgeGraph
    )

Example usage

graph = generate_knowledge_graph("Quantum mechanics and its applications")

Print the nodes and edges

for node in graph.nodes:
    print(f"Node {node.id}: {node.label} ({node.color})")

for edge in graph.edges:
    print(f"Edge: {edge.source} --({edge.label})--> {edge.target}")

To visualize the knowledge graph, you can use libraries like graphviz:

from graphviz import Digraph

def visualize_knowledge_graph(kg: KnowledgeGraph):
    dot = Digraph(comment="Knowledge Graph")

    # Add nodes
    for node in kg.nodes:
        dot.node(str(node.id), node.label, color=node.color)

    # Add edges
    for edge in kg.edges:
        dot.edge(str(edge.source), str(edge.target),
                 label=edge.label, color=edge.color)

    # Render the graph
    dot.render("knowledge_graph.gv", view=True)

Visualize the graph

visualize_knowledge_graph(graph)

Running the Example

First, install Instructor and any dependencies

$ pip install instructor pydantic

Run the Python script

$ python knowledge-graphs.py

Further Information

Documentation link 1