{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "746ccb3d604cd78d",
   "metadata": {},
   "source": [
    "# Knowledge Graph\n",
    "\n",
    "Often times, data is messy and not always stored in a structured manner ready for use by an LLM. In this recipe, we show how to create a knowledge graph from an unstructured document using common python libraries and Mirascope using OpenAI GPT-4o-mini.\n",
    "\n",
    "<div class=\"admonition tip\">\n",
    "<p class=\"admonition-title\">Mirascope Concepts Used</p>\n",
    "<ul>\n",
    "<li><a href=\"../../../learn/prompts/\">Prompts</a></li>\n",
    "<li><a href=\"../../../learn/calls/\">Calls</a></li>\n",
    "<li><a href=\"../../../learn/response_models/\">Response Models</a></li>\n",
    "</ul>\n",
    "</div>\n",
    "\n",
    "<div class=\"admonition note\">\n",
    "<p class=\"admonition-title\">Background</p>\n",
    "<p>\n",
    "While traditional Natural Language Processing (NLP) techniques have long been used in knowledge graphs to identify entities and relationships in unstructured text, Large Language Models (LLMs) have significantly improved this process. LLMs enhance the accuracy of entity identification and linking to knowledge graph entries, demonstrating superior ability to handle context and ambiguity compared to conventional NLP methods. \n",
    "</p>\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cf302964",
   "metadata": {},
   "source": [
    "## Setup\n",
    "\n",
    "To set up our environment, first let's install all of the packages we will use:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c7e5561d",
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install \"mirascope[openai]\"\n",
    "# (Optional) For visualization\n",
    "!pip install matplotlib networkx\n",
    "# (Optional) For parsing HTML\n",
    "!pip install beautifulsoup4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "19f1cad4",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "os.environ[\"OPENAI_API_KEY\"] = \"YOUR_API_KEY\"\n",
    "# Set the appropriate API key for the provider you're using"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2792e13e4196e11c",
   "metadata": {},
   "source": [
    "\n",
    "## Create the `KnowledgeGraph`\n",
    "\n",
    "The first step is to create a `KnowledgeGraph` with `Nodes` and `Edges` that represent our entities and relationships. For our simple recipe, we will use a Pydantic `BaseModel` to represent our `KnowledgeGraph`:\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a661e68462bd8cf9",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T08:44:11.812873Z",
     "start_time": "2024-09-30T08:44:11.741113Z"
    }
   },
   "outputs": [],
   "source": [
    "from pydantic import BaseModel, Field\n",
    "\n",
    "\n",
    "class Edge(BaseModel):\n",
    "    source: str = Field(..., description=\"The source node of the edge\")\n",
    "    target: str = Field(..., description=\"The target node of the edge\")\n",
    "    relationship: str = Field(\n",
    "        ..., description=\"The relationship between the source and target nodes\"\n",
    "    )\n",
    "\n",
    "\n",
    "class Node(BaseModel):\n",
    "    id: str = Field(..., description=\"The unique identifier of the node\")\n",
    "    type: str = Field(..., description=\"The type or label of the node\")\n",
    "    properties: dict | None = Field(\n",
    "        ..., description=\"Additional properties and metadata associated with the node\"\n",
    "    )\n",
    "\n",
    "\n",
    "class KnowledgeGraph(BaseModel):\n",
    "    nodes: list[Node] = Field(..., description=\"List of nodes in the knowledge graph\")\n",
    "    edges: list[Edge] = Field(..., description=\"List of edges in the knowledge graph\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "60d778e6f10f5198",
   "metadata": {},
   "source": [
    "\n",
    "Our `Edge` represents connections between nodes, with attributes for the source node, target node, and the relationship between them. While our `Node` defines nodes with an ID, type, and optional properties. Our `KnowledgeGraph` then aggregates these nodes and edges into a comprehensive knowledge graph.\n",
    "\n",
    "Now that we have our schema defined, it's time to create our knowledge graph.\n",
    "\n",
    "## Creating the knowledge graph\n",
    "\n",
    "We start off with engineering our prompt, prompting the LLM to create a knowledge graph based on the user query. Then we are taking a [Wikipedia](https://en.wikipedia.org/wiki/Large_language_model) article and converting the raw text into a structured knowledge graph.\n",
    "The command below will download the article to your local machine by using the `curl` command. If you don't have `curl` installed, you can download the article manually from the link above and save it as `wikipedia.html`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "222fdaa0fac9b4a",
   "metadata": {},
   "outputs": [],
   "source": [
    "!curl https://en.wikipedia.org/wiki/Large_language_model -o wikipedia.html"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "35677693",
   "metadata": {},
   "source": [
    "{% raw %}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "b1ebdb2c25405585",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T08:45:20.988369Z",
     "start_time": "2024-09-30T08:45:08.989406Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "nodes=[Node(id='Large Language Models', type='Large Language Model', properties=None), Node(id='Data Cleaning Issues', type='Pitfall', properties=None), Node(id='Bias Inheritance', type='Pitfall', properties=None), Node(id='Hallucinations', type='Pitfall', properties=None), Node(id='Limited Understanding', type='Pitfall', properties=None), Node(id='Dependence on Training Data', type='Pitfall', properties=None), Node(id='Security Risks', type='Pitfall', properties=None), Node(id='Stereotyping', type='Pitfall', properties=None), Node(id='Political Bias', type='Pitfall', properties=None)] edges=[Edge(source='Large Language Models', target='Data Cleaning Issues', relationship='has pitfall'), Edge(source='Large Language Models', target='Bias Inheritance', relationship='has pitfall'), Edge(source='Large Language Models', target='Hallucinations', relationship='has pitfall'), Edge(source='Large Language Models', target='Limited Understanding', relationship='has pitfall'), Edge(source='Large Language Models', target='Dependence on Training Data', relationship='has pitfall'), Edge(source='Large Language Models', target='Security Risks', relationship='has pitfall'), Edge(source='Large Language Models', target='Stereotyping', relationship='has pitfall'), Edge(source='Large Language Models', target='Political Bias', relationship='has pitfall')]\n"
     ]
    }
   ],
   "source": [
    "from bs4 import BeautifulSoup\n",
    "from mirascope.core import openai, prompt_template\n",
    "\n",
    "\n",
    "def get_text_from_html(file_path: str) -> str:\n",
    "    with open(file_path) as file:\n",
    "        html_text = file.read()\n",
    "    return BeautifulSoup(html_text, \"html.parser\").get_text()\n",
    "\n",
    "\n",
    "@openai.call(model=\"gpt-4o-mini\", response_model=KnowledgeGraph)\n",
    "@prompt_template(\n",
    "    \"\"\"\n",
    "    SYSTEM:\n",
    "    Your job is to create a knowledge graph based on the text and user question.\n",
    "    \n",
    "    The article:\n",
    "    {text}\n",
    "\n",
    "    Example:\n",
    "    John and Jane Doe are siblings. Jane is 25 and 5 years younger than John.\n",
    "    Node(id=\"John Doe\", type=\"Person\", properties={{\"age\": 30}})\n",
    "    Node(id=\"Jane Doe\", type=\"Person\", properties={{\"age\": 25}})\n",
    "    Edge(source=\"John Doe\", target=\"Jane Doe\", relationship=\"Siblings\")\n",
    "\n",
    "    USER:\n",
    "    {question}\n",
    "    \"\"\"\n",
    ")\n",
    "def generate_knowledge_graph(\n",
    "    question: str, file_name: str\n",
    ") -> openai.OpenAIDynamicConfig:\n",
    "    text = get_text_from_html(file_name)\n",
    "    return {\"computed_fields\": {\"text\": text}}\n",
    "\n",
    "\n",
    "question = \"What are the pitfalls of using LLMs?\"\n",
    "\n",
    "kg = generate_knowledge_graph(question, \"wikipedia.html\")\n",
    "print(kg)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "be5cd8d7",
   "metadata": {},
   "source": [
    "{% endraw %}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b4340f0cf9bbb98b",
   "metadata": {},
   "source": [
    "We engineer our prompt by giving examples of how the properties should be filled out and use Mirascope's `DynamicConfig` to pass in the article. While it seems silly in this context, there may be multiple documents that you may want to conditionally pass in depending on the query. This can include text chunks from a Vector Store or data from a Database.\n",
    "\n",
    "After we generated our knowledge graph, it is time to create our `run` function\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "fa188bc9306f1342",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T08:51:51.291953Z",
     "start_time": "2024-09-30T08:51:51.285559Z"
    }
   },
   "outputs": [],
   "source": [
    "@openai.call(model=\"gpt-4o-mini\")\n",
    "@prompt_template(\n",
    "    \"\"\"\n",
    "    SYSTEM:\n",
    "    Answer the following question based on the knowledge graph.\n",
    "\n",
    "    Knowledge Graph:\n",
    "    {knowledge_graph}\n",
    "    \n",
    "    USER:\n",
    "    {question}\n",
    "    \"\"\"\n",
    ")\n",
    "def run(question: str, knowledge_graph: KnowledgeGraph): ..."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a09945fb1fce8d5b",
   "metadata": {},
   "source": [
    "We define a simple `run` function that answers the users query based on the knowledge graph. Combining knowledge graphs with semantic search will lead to the LLM having better context to address complex questions.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "2387b908e882363a",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T08:52:42.452370Z",
     "start_time": "2024-09-30T08:52:39.914507Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The pitfalls of using Large Language Models (LLMs) include:\n",
      "\n",
      "1. Data Cleaning Issues\n",
      "2. Bias Inheritance\n",
      "3. Hallucinations\n",
      "4. Limited Understanding\n",
      "5. Dependence on Training Data\n",
      "6. Security Risks\n",
      "7. Stereotyping\n",
      "8. Political Bias\n"
     ]
    }
   ],
   "source": [
    "print(run(question, kg))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8e8d6eaf232a8a85",
   "metadata": {},
   "source": [
    "## Render your graph\n",
    "\n",
    "Optionally, to visualize the knowledge graph, we use networkx and matplotlib to draw the edges and nodes.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "86fbc75810894c93",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-09-30T08:53:48.312430Z",
     "start_time": "2024-09-30T08:53:38.101837Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Matplotlib is building the font cache; this may take a moment.\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import networkx as nx\n",
    "\n",
    "\n",
    "def render_graph(kg: KnowledgeGraph):\n",
    "    G = nx.DiGraph()\n",
    "\n",
    "    for node in kg.nodes:\n",
    "        G.add_node(node.id, label=node.type, **(node.properties or {}))\n",
    "\n",
    "    for edge in kg.edges:\n",
    "        G.add_edge(edge.source, edge.target, label=edge.relationship)\n",
    "\n",
    "    plt.figure(figsize=(15, 10))\n",
    "    pos = nx.spring_layout(G)\n",
    "\n",
    "    nx.draw_networkx_nodes(G, pos, node_size=2000, node_color=\"lightblue\")\n",
    "    nx.draw_networkx_edges(G, pos, arrowstyle=\"->\", arrowsize=20)\n",
    "    nx.draw_networkx_labels(G, pos, font_size=12, font_weight=\"bold\")\n",
    "\n",
    "    edge_labels = nx.get_edge_attributes(G, \"label\")\n",
    "    nx.draw_networkx_edge_labels(G, pos, edge_labels=edge_labels, font_color=\"red\")\n",
    "\n",
    "    plt.title(\"Knowledge Graph Visualization\", fontsize=15)\n",
    "    plt.show()\n",
    "\n",
    "\n",
    "question = \"What are the pitfalls of using LLMs?\"\n",
    "render_graph(kg)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "281bc33a958300c5",
   "metadata": {},
   "source": [
    "<div class=\"admonition tip\">\n",
    "<p class=\"admonition-title\">Additional Real-World Applications</p>\n",
    "<ol>\n",
    "<li><p><b>Enhance your Q&A</b></p>\n",
    "<ul>\n",
    "<li>Customer support system uses knowledge graph containing information about products to answer questions.</li>\n",
    "<li>Example: \"Does the Mirascope phone support fast charging?\" The knowledge graph has a node \"Mirascope smartphone\" and searches \"support\" edge to find fast charging and returns results for the LLM to use.</li>\n",
    "</ul>\n",
    "</li>\n",
    "<li><p><b>Supply Chain Optimization</b></p>\n",
    "<ul>\n",
    "<li>A knowledge graph could represent complex relationships between suppliers, manufacturing plants, distribution centers, products, and transportation routes.</li>\n",
    "<li>Example: How would a 20% increase in demand for a mirascope affect our inventory needs and shipping costs? Use knowledge graph to trace the mirascope toy, calculate inventory, and then estimate shipping costs and return results for the LLM to give a report.</li>\n",
    "</ul>\n",
    "</li>\n",
    "<li><p><b>Healthcare Assistant</b></p>\n",
    "<ul>\n",
    "<li>Assuming no PII or HIPPA violation, build a knowledge graph from patient remarks.</li>\n",
    "<li>Example: \"Mary said help, I've fallen\". Build up a knowledge graph from comments and use an LLM to scan the node \"Mary\" for any worrying activity. Have the LLM alert Healthcare employees that there may be an emergency.</li>\n",
    "</ul>\n",
    "</li>\n",
    "</ol>\n",
    "</div>\n",
    "\n",
    "When adapting this recipe, consider:\n",
    "\n",
    "- Combining knowledge graph with Text Embeddings for both structured search and semantic search, depending on your requirements.\n",
    "- Store your knowledge graph in a database / cache for faster retrieval.\n",
    "- Experiment with different LLM models, some may be better than others for generating the knowledge graph.\n",
    "- Turn the example into an Agentic workflow, giving it access to tools such as web search so the LLM can call tools to update its own knowledge graph to answer any question.\n",
    "- Adding Pydantic `AfterValidators` to prevent duplicate Node IDs.\n",
    "\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}