Abruf aus dem Graphen verbessern

Retrieval Augmented Generation (RAG) mit LangChain

Meri Nova

Machine Learning Engineer

Techniken

Hauptgrenze: Zuverlässigkeit der Übersetzung Nutzerfrage → Cypher

Strategien zur Verbesserung des Graph-Retrieval-Systems:

Graph-Schema filtern
Cypher-Query validieren
Few-shot Prompting

Filtern

from langchain_community.chains.graph_qa.cypher import GraphCypherQAChain

llm = ChatOpenAI(api_key="...", model="gpt-4o-mini", temperature=0)


chain = GraphCypherQAChain.from_llm(

    graph=graph, llm=llm, exclude_types=["Concept"], verbose=True

)

print(graph.get_schema)

Node properties:
Document {title: STRING, id: STRING, text: STRING, summary: STRING, source: STRING}
Organization {id: STRING}

Cypher-Query validieren

Schwierigkeit, die Richtung von Beziehungen zu interpretieren

chain = GraphCypherQAChain.from_llm(
    graph=graph, llm=llm, verbose=True, validate_cypher=True
)

Ermittelt Knoten und Beziehungen
Bestimmt die Richtungen der Beziehungen
Prüft das Graph-Schema
Korrigiert die Beziehungsrichtungen

Few-shot Prompting

examples = [
    {
        "question": "How many notable large language models are mentioned in the article?",
        "query": "MATCH (m:Concept {id: 'Large Language Model'}) RETURN count(DISTINCT m)",
    },
    {
        "question": "Which companies or organizations have developed the large language models mentioned?",
        "query": "MATCH (o:Organization)-[:DEVELOPS]->(m:Concept {id: 'Large Language Model'}) RETURN DISTINCT o.id",
    },
    {
        "question": "What is the largest model size mentioned in the article, in terms of number of parameters?",
        "query": "MATCH (m:Concept {id: 'Large Language Model'}) RETURN max(m.parameters) AS largest_model",
    },
]

Few-shot Prompting umsetzen

from langchain_core.prompts import FewShotPromptTemplate, PromptTemplate


example_prompt = PromptTemplate.from_template(
    "User input: {question}\nCypher query: {query}"
)


cypher_prompt = FewShotPromptTemplate(
    examples=examples,
    example_prompt=example_prompt,
    prefix="You are a Neo4j expert. Given an input question, create a syntactically correct
    Cypher query to run.\n\nHere is the schema information\n{schema}.\n\n
    Below are a number of examples of questions and their corresponding Cypher queries.",
    suffix="User input: {question}\nCypher query: ",
    input_variables=["question"],
)

Vollständiger Prompt

Du bist Neo4j-Expert:in. Erstelle zu einer Eingabefrage eine syntaktisch korrekte Cypher-Query.

Unten stehen Beispiele für Fragen und die zugehörigen Cypher-Queries.

User input: How many notable large language models are mentioned in the article?
Cypher query: MATCH (p:Paper) RETURN count(DISTINCT p)

User input: Which companies or organizations have developed the large language models?
Cypher query: MATCH (o:Organization)-[:DEVELOPS]->(m:Concept {id: 'Large Language Model'}) RETURN DISTINCT o.id

User input: What is the largest model size mentioned in the article, in terms of number of parameters?
Cypher query: MATCH (m:Concept {id: 'Large Language Model'}) RETURN max(m.parameters) AS largest_model

User input: How many papers were published in 2016?
Cypher query:

Few-shot-Beispiele hinzufügen

chain = GraphCypherQAChain.from_llm(
    graph=graph, llm=llm, cypher_prompt=cypher_prompt,
    verbose=True, validate_cypher=True
)

Lass uns üben!

Retrieval Augmented Generation (RAG) mit LangChain