使用 PDF 向量、OCR、GPT-4 和 Google Drive 的研究论文分析系统

• •OpenAI API Key
• •PostgreSQL 数据库连接信息
• •Google Drive API 凭证

{
  "meta": {
    "instanceId": "placeholder"
  },
  "nodes": [
    {
      "id": "overview-note",
      "name": "研究概览",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        50,
        50
      ],
      "parameters": {
        "color": 5,
        "width": 350,
        "height": 180,
        "content": "## 📚 研究论文分析器"
      },
      "typeVersion": 1
    },
    {
      "id": "search-note",
      "name": "学术搜索",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        450,
        450
      ],
      "parameters": {
        "width": 260,
        "height": 150,
        "content": "## 🔍 论文搜索"
      },
      "typeVersion": 1
    },
    {
      "id": "extract-note",
      "name": "论文提取",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        750,
        450
      ],
      "parameters": {
        "width": 260,
        "height": 180,
        "content": "## 📄 提取"
      },
      "typeVersion": 1
    },
    {
      "id": "summary-note",
      "name": "AI 分析",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        1050,
        450
      ],
      "parameters": {
        "color": 6,
        "width": 260,
        "height": 180,
        "content": "## 🤖 AI 摘要"
      },
      "typeVersion": 1
    },
    {
      "id": "manual-trigger",
      "name": "手动触发器",
      "type": "n8n-nodes-base.manualTrigger",
      "notes": "Start paper analysis",
      "position": [
        250,
        300
      ],
      "parameters": {},
      "typeVersion": 1
    },
    {
      "id": "google-drive",
      "name": "Google Drive - 获取论文",
      "type": "n8n-nodes-base.googleDrive",
      "notes": "Retrieve paper from Drive",
      "position": [
        450,
        300
      ],
      "parameters": {
        "fileId": "={{ $json.fileId }}",
        "operation": "download"
      },
      "typeVersion": 3
    },
    {
      "id": "pdfvector-parse",
      "name": "PDF Vector - 解析论文",
      "type": "n8n-nodes-pdfvector.pdfVector",
      "notes": "Parse research paper",
      "position": [
        650,
        300
      ],
      "parameters": {
        "useLLM": "always",
        "resource": "document",
        "inputType": "file",
        "operation": "parse",
        "binaryPropertyName": "data"
      },
      "typeVersion": 1
    },
    {
      "id": "pdfvector-extract",
      "name": "PDF Vector - 提取数据",
      "type": "n8n-nodes-pdfvector.pdfVector",
      "notes": "Extract structured data",
      "position": [
        850,
        300
      ],
      "parameters": {
        "prompt": "Extract key information from this research document or image including title, authors with affiliations, abstract, keywords, research questions, methodology, key findings, conclusions, limitations, and future work suggestions. Use OCR if this is a scanned document or image.",
        "schema": "{\"type\":\"object\",\"properties\":{\"title\":{\"type\":\"string\"},\"authors\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"properties\":{\"name\":{\"type\":\"string\"},\"affiliation\":{\"type\":\"string\"},\"email\":{\"type\":\"string\"}}}},\"abstract\":{\"type\":\"string\"},\"keywords\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"researchQuestions\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"methodology\":{\"type\":\"object\",\"properties\":{\"approach\":{\"type\":\"string\"},\"dataCollection\":{\"type\":\"string\"},\"analysis\":{\"type\":\"string\"},\"tools\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}}}},\"findings\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"conclusions\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"limitations\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"futureWork\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}},\"references\":{\"type\":\"number\"}},\"required\":[\"title\",\"authors\"],\"additionalProperties\":false}",
        "resource": "document",
        "inputType": "file",
        "operation": "extract",
        "binaryPropertyName": "data"
      },
      "typeVersion": 1
    },
    {
      "id": "openai-analyze",
      "name": "生成 AI 摘要",
      "type": "n8n-nodes-base.openAi",
      "notes": "Create AI summary",
      "position": [
        1050,
        300
      ],
      "parameters": {
        "model": "gpt-4",
        "messages": {
          "values": [
            {
              "content": "Based on this research paper, provide:\n\n1. A concise summary (150 words) suitable for a research database\n2. The main contribution to the field (2-3 sentences)\n3. Potential applications or impact\n4. Classification tags (e.g., empirical study, theoretical framework, review, etc.)\n\nPaper content:\n{{ $node['PDF Vector - Parse Paper'].json.content }}"
            }
          ]
        }
      },
      "typeVersion": 1
    },
    {
      "id": "prepare-data",
      "name": "准备数据库条目",
      "type": "n8n-nodes-base.code",
      "notes": "Combine all data",
      "position": [
        1250,
        300
      ],
      "parameters": {
        "jsCode": "// Combine all analysis data\nconst parsedContent = $node['PDF Vector - Parse Paper'].json;\nconst extractedData = $node['PDF Vector - Extract Data'].json.data;\nconst aiSummary = $node['Generate AI Summary'].json.choices[0].message.content;\n\n// Calculate reading time (assuming 250 words per minute)\nconst wordCount = parsedContent.content.split(' ').length;\nconst readingTimeMinutes = Math.ceil(wordCount / 250);\n\n// Prepare database entry\nconst paperAnalysis = {\n  // Basic information\n  title: extractedData.title,\n  authors: extractedData.authors,\n  url: $node['Google Drive - Get Paper'].json.webViewLink,\n  \n  // Content\n  abstract: extractedData.abstract,\n  keywords: extractedData.keywords,\n  fullText: parsedContent.content,\n  \n  // Analysis\n  aiSummary: aiSummary,\n  methodology: extractedData.methodology,\n  findings: extractedData.findings,\n  conclusions: extractedData.conclusions,\n  limitations: extractedData.limitations,\n  futureWork: extractedData.futureWork,\n  \n  // Metadata\n  wordCount: wordCount,\n  readingTimeMinutes: readingTimeMinutes,\n  referenceCount: extractedData.references || 0,\n  processedAt: new Date().toISOString(),\n  \n  // Searchable fields\n  searchText: `${extractedData.title} ${extractedData.abstract} ${extractedData.keywords.join(' ')}`.toLowerCase()\n};\n\nreturn [{ json: paperAnalysis }];"
      },
      "typeVersion": 2
    },
    {
      "id": "database-store",
      "name": "存储到数据库",
      "type": "n8n-nodes-base.postgres",
      "notes": "Save to research database",
      "position": [
        1450,
        300
      ],
      "parameters": {
        "table": "research_papers",
        "columns": "title,authors,url,abstract,keywords,ai_summary,methodology,findings,processed_at,search_text",
        "operation": "insert"
      },
      "typeVersion": 1
    }
  ],
  "connections": {
    "Manual Trigger": {
      "main": [
        [
          {
            "node": "Google Drive - Get Paper",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Generate AI Summary": {
      "main": [
        [
          {
            "node": "Prepare Database Entry",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Prepare Database Entry": {
      "main": [
        [
          {
            "node": "Store in Database",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Google Drive - Get Paper": {
      "main": [
        [
          {
            "node": "PDF Vector - Parse Paper",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "PDF Vector - Parse Paper": {
      "main": [
        [
          {
            "node": "PDF Vector - Extract Data",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "PDF Vector - Extract Data": {
      "main": [
        [
          {
            "node": "Generate AI Summary",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  }
}