使用llama-index连接neo4j知识图谱 达成大模型构建查询知识图谱功能

使用llama-index连接neo4j知识图谱 达成本地大模型构建查询知识图谱功能

概述

目标

我们的目标是使用Llama-Index来连接到Neo4j，以构建和查询知识图谱。通过这个过程，我们能够将文档中的信息转化为知识图谱，并通过大语言模型进行查询。
以下参考llama-index官方实现

主要步骤

1. 安装依赖项

首先，我们需要安装一些Python库。这些库包括Llama-Index的相关组件和Neo4j的连接库。

%pip install llama-index-llms-openai
%pip install llama-index-graph-stores-neo4j
%pip install llama-index-embeddings-openai
%pip install llama-index-llms-azure-openai
%pip install neo4j

这些库的功能如下：

2. 配置环境

为了使用OpenAI或Azure OpenAI的API，我们需要配置一些环境变量和API密钥。

配置ollama

import os
from llama_index.core import VectorStoreIndex, SimpleDirectoryReader, Settings, StorageContext, KnowledgeGraphIndex
from llama_index.embeddings.ollama import OllamaEmbedding
from llama_index.llms.ollama import Ollama
from llama_index.core import KnowledgeGraphIndex, SimpleDirectoryReader, StorageContext
from llama_index.graph_stores.neo4j import Neo4jGraphStore

# 设置嵌入模型
Settings.embed_model = OllamaEmbedding(model_name="znbang/bge:large-zh-v1.5-f32")

# 设置LLM模型
Settings.llm = Ollama(model="qwen:7b", request_timeout=360.0

3. 使用Neo4j构建知识图谱

准备Neo4j

我们需要配置Neo4j数据库的连接信息。

username = "neo4j"
password = "your-neo4j-password"
url = "bolt://your-neo4j-url:7687"
database = "neo4j"

实例化Neo4jGraph KG索引

接下来，我们使用Llama-Index从文档中提取数据，并将其存储到Neo4j图数据库中。

from llama_index.core import KnowledgeGraphIndex, SimpleDirectoryReader
from llama_index.core import StorageContext
from llama_index.graph_stores.neo4j import Neo4jGraphStore

from llama_index.llms.openai import OpenAI
from IPython.display import Markdown, display

# 加载文档数据
documents = SimpleDirectoryReader(
    "path_to_your_documents"
).load_data()

# 初始化Neo4j图存储
graph_store = Neo4jGraphStore(
    username=username,
    password=password,
    url=url,
    database=database,
)

# 创建存储上下文
storage_context = StorageContext.from_defaults(graph_store=graph_store)

# 构建知识图谱索引
index = KnowledgeGraphIndex.from_documents(
    documents,
    storage_context=storage_context,
    max_triplets_per_chunk=2,
)

4. 查询知识图谱

查询并仅发送三元组到LLM

我们可以查询知识图谱并仅发送三元组到大语言模型进行处理。

query_engine = index.as_query_engine(
    include_text=False, response_mode="tree_summarize"
)

response = query_engine.query("Tell me more about Interleaf")
display(Markdown(f"<b>{response}</b>"))

查询并发送包含文本的结果到LLM

我们还可以查询知识图谱并发送包含文本的结果到大语言模型。

query_engine = index.as_query_engine(
    include_text=True, response_mode="tree_summarize"
)
response = query_engine.query(
    "Tell me more about what the author worked on at Interleaf"
)
display(Markdown(f"<b>{response}</b>"))

5. 使用嵌入查询

我们还可以在构建索引时包含文本嵌入，以便在查询时使用嵌入相似度进行更准确的查询。

# 清理数据集
graph_store.query(
    """
MATCH (n) DETACH DELETE n
"""
)

# 构建包含嵌入的索引
index = KnowledgeGraphIndex.from_documents(
    documents,
    storage_context=storage_context,
    max_triplets_per_chunk=2,
    include_embeddings=True,
)

query_engine = index.as_query_engine(
    include_text=True,
    response_mode="tree_summarize",
    embedding_mode="hybrid",
    similarity_top_k=5,
)

response = query_engine.query(
    "Tell me more about what the author worked on at Interleaf"
)
display(Markdown(f"<b>{response}</b>"))

6. 手动添加三元组并构建图谱（可选）

我们还可以手动向知识图谱中添加三元组。

from llama_index.core.node_parser import SentenceSplitter
node_parser = SentenceSplitter()
nodes = node_parser.get_nodes_from_documents(documents)

# 初始化一个空的索引
index = KnowledgeGraphIndex.from_documents([], storage_context=storage_context)

# 手动添加三元组
node_0_tups = [
    ("author", "worked on", "writing"),
    ("author", "worked on", "programming"),
]
for tup in node_0_tups:
    index.upsert_triplet_and_node(tup, nodes[0])

node_1_tups = [
    ("Interleaf", "made software for", "creating documents"),
    ("Interleaf", "added", "scripting language"),
    ("software", "generate", "web sites"),
]
for tup in node_1_tups:
    index.upsert_triplet_and_node(tup, nodes[1])

query_engine = index.as_query_engine(
    include_text=False, response_mode="tree_summarize"
)

response = query_engine.query("Tell me more about Interleaf")
display(Markdown(f"<b>{response}</b>"))