{
  "schema_version": "2.0",
  "record_type": "dataset",
  "tier": "verified",
  "slug": "meddialog-cn",
  "name": "MedDialog-CN",
  "summary": {
    "zh": "许可证声明：无法核实（NOASSERTION）；本站处理：待确认，暂只链接原出处。许可证无法核实：唯一的许可信息来自无法证明属于发布方的 Hugging Face 页面（按第三方镜像记为二手），没有一手声明；确认前只链接原出处。",
    "en": "License: unverifiable (NOASSERTION); Mdexa hosting: pending. License unverifiable: The only license information is on a Hugging Face page whose account cannot be shown to belong to the publisher (third-party mirror, secondhand); no first-hand statement; link only until resolved."
  },
  "systems": [
    "modern-medicine"
  ],
  "domains": [
    "clinical-medicine"
  ],
  "content_types": [
    "qa-dialogue"
  ],
  "content_description": {
    "en": "Doctor-patient dialogue corpus",
    "zh": "医患对话语料"
  },
  "languages": [
    "zh"
  ],
  "modalities": [
    "text"
  ],
  "size": {
    "raw": "论文称中文340万条对话；HF 镜像 raw zh 1,921,127条，processed.zh 约340万条",
    "measures": [
      {
        "value": 3400000,
        "unit": "dialogues",
        "qualifier": "claimed",
        "scope": "paper, Chinese"
      },
      {
        "value": 1921127,
        "unit": "dialogues",
        "qualifier": "exact",
        "scope": "HF mirror raw zh"
      },
      {
        "value": 3400000,
        "unit": "dialogues",
        "qualifier": "approx",
        "scope": "HF mirror processed.zh"
      }
    ]
  },
  "publisher": "UCSD-AI4H（EMNLP 2020 论文）",
  "year": {
    "value": 2020,
    "raw": "2020"
  },
  "expert_review": {
    "en": "None",
    "zh": "无"
  },
  "license": {
    "legacy_status": "index-only",
    "spdx": "NOASSERTION",
    "commercial_use": "unverified",
    "derivatives": "unverified",
    "redistribution": "unverified",
    "rationale": {
      "en": "The only license information is on a Hugging Face page whose account cannot be shown to belong to the publisher (third-party mirror, secondhand); no first-hand statement",
      "zh": "唯一的许可信息来自无法证明属于发布方的 Hugging Face 页面（按第三方镜像记为二手），没有一手声明"
    },
    "evidence": [
      {
        "location": "hf-card-metadata",
        "location_detail": {
          "en": "Hugging Face dataset card license field",
          "zh": "Hugging Face 数据集卡 license 字段"
        },
        "excerpt": "unknown",
        "spdx": "NOASSERTION",
        "url": "https://huggingface.co/datasets/UCSD26/medical_dialog",
        "verified_on": "2026-11-12",
        "snapshot_id": "s01",
        "observed": "page",
        "secondhand": true
      },
      {
        "location": "hf-card-text",
        "location_detail": {
          "en": "Dataset card copyright statement",
          "zh": "数据集卡版权声明"
        },
        "excerpt": "All copyrights of the data belong to haodf.com.",
        "spdx": null,
        "url": "https://huggingface.co/datasets/UCSD26/medical_dialog",
        "verified_on": "2026-11-12",
        "snapshot_id": "s01",
        "observed": "page",
        "secondhand": true
      }
    ]
  },
  "license_status": "unverifiable",
  "hosting": "pending",
  "hosting_reason": {
    "zh": "许可证无法核实：唯一的许可信息来自无法证明属于发布方的 Hugging Face 页面（按第三方镜像记为二手），没有一手声明；确认前只链接原出处。",
    "en": "License unverifiable: The only license information is on a Hugging Face page whose account cannot be shown to belong to the publisher (third-party mirror, secondhand); no first-hand statement; link only until resolved."
  },
  "signoff": null,
  "risk_flags": [
    "scraped",
    "real-patient-text"
  ],
  "provenance": {
    "source_description": {
      "en": "Scraped from haodf.com (好大夫在线)",
      "zh": "爬取自好大夫在线（haodf.com）"
    },
    "upstream": [
      {
        "slug": null,
        "name": "haodf.com (好大夫在线)",
        "relation": "based-on"
      }
    ]
  },
  "known_issues": [
    {
      "en": "Copyright belongs to the source website",
      "zh": "版权归来源网站"
    },
    {
      "en": "The card notes that the English source site's terms prohibit scraping",
      "zh": "卡片注明英文来源站点条款禁止抓取"
    },
    {
      "en": "GitHub repository UCSD-AI4H/Medical-Dialogue-System returned 404 this time",
      "zh": "GitHub 仓库地址 UCSD-AI4H/Medical-Dialogue-System 本次返回404"
    },
    {
      "en": "Source risk: dialogues were scraped from haodf.com (website scraping) and are real doctor-patient conversations (real patient text)",
      "zh": "来源风险：对话抓取自好大夫在线（网站抓取），且为真实医患对话（真实患者文本）"
    },
    {
      "en": "The Hugging Face account hosting the data cannot be shown to be the publisher; its license information is treated as secondhand",
      "zh": "托管数据的 Hugging Face 账号无法证明是发布方，其许可信息按二手处理"
    }
  ],
  "preflight": {
    "status": "not-run"
  },
  "access": {
    "mdexa_download": null,
    "external_urls": [
      "https://huggingface.co/datasets/UCSD26/medical_dialog"
    ],
    "notes": null
  },
  "citation": {
    "text": null,
    "bibtex": null,
    "references": []
  },
  "verification_level": "verified",
  "last_verified": "2026-11-12",
  "tags": [
    {
      "facet": "task_format",
      "value": "multi-turn-dialogue",
      "basis": "author-stated",
      "excerpt": "- Dialogue : The conversation between the doctor and the patient.",
      "snapshot_id": "s01"
    },
    {
      "facet": "official_split",
      "value": "full-split",
      "basis": "author-stated",
      "excerpt": "For processed configurations, data is split into train, validation and test, with the following number of examples:",
      "snapshot_id": "s01",
      "scope": {
        "en": "processed configurations",
        "zh": "processed 配置"
      }
    },
    {
      "facet": "specialty",
      "value": "multi-specialty",
      "basis": "inferred",
      "inferred_from": "card-content-description",
      "note": {
        "zh": "综合医患对话，未限定专科",
        "en": "综合医患对话; not limited to a specialty"
      }
    }
  ],
  "changelog": [
    {
      "date": "2026-11-12",
      "change": {
        "en": "Card converted from an earlier manual catalog; no new facts added; reclassified under taxonomy v2.",
        "zh": "由早期人工整理的目录转换生成，未新增事实；按分类体系 v2 重新归类。"
      }
    },
    {
      "date": "2026-11-12",
      "change": {
        "en": "License status split into license_status (what the license says) and hosting (what Mdexa does); the earlier conclusion is kept for traceability.",
        "zh": "把许可证状态拆成 license_status（许可证怎么写）与 hosting（本站怎么处理）；原结论保留以便追溯。"
      }
    },
    {
      "date": "2026-11-12",
      "change": {
        "en": "Added verbatim license excerpts and snapshot ids from captured page snapshots; no conclusion changed.",
        "zh": "补入来自已抓取页面快照的许可证逐字摘录与快照编号；未改动任何结论。"
      }
    },
    {
      "date": "2026-11-12",
      "change": {
        "en": "License status changed from restricted to unverifiable and hosting from link-only to pending: third-party mirror pages are secondhand; secondhand evidence alone cannot support restricted.",
        "zh": "许可状态由受限改为无法核实，处理方式由只链接原出处改为待确认：第三方镜像页记为二手，二手证据不能单独支撑 restricted。"
      }
    },
    {
      "date": "2026-11-13",
      "change": {
        "en": "Added dataset-level tags: 3 (2 stated by the authors with excerpts, 1 inferred from the content). No conclusion changed.",
        "zh": "新增数据集标签 3 个（作者声明 2 个，附逐字摘录；据内容推断 1 个）；未改动任何结论。"
      }
    },
    {
      "date": "2026-11-13",
      "change": {
        "en": "Removed the stored size-tier tag; the tier is now computed from the size field when pages are built.",
        "zh": "移除存储的规模档位标签；档位改为生成页面时由 size 字段计算。"
      }
    },
    {
      "date": "2026-11-13",
      "change": {
        "en": "Added specialty tags: multi-specialty(inferred).",
        "zh": "新增专科标签：multi-specialty(inferred)。"
      }
    }
  ],
  "_links": {
    "text": "在本库继续查找。本条目是 109 条已处理并核实的记录之一（数据集、数据资源与开源模型）。查找其他条目：调用 `https://mcp.mdexa.com/v1/datasets?q=<关键词>`，或连接 MCP 服务 `https://mcp.mdexa.com/mcp`；每分钟 20 次以内无需密钥。每行返回不到 1 KB；空结果即\"数据库未收录\"。核对一份清单：一次批量核对（免费账号）即返回报告。说明：/agents.md。相似条目：CMtMedQA (https://mdexa.com/zh/datasets/cmtmedqa/)；DISC-Med-SFT (https://mdexa.com/zh/datasets/disc-med-sft/)；Huatuo-Llama-Med-Chinese 知识库问答与文献数据 (https://mdexa.com/zh/datasets/huatuo-llama-med-chinese-data/)。",
    "records": 109,
    "search": {
      "template": "https://mcp.mdexa.com/v1/datasets?q={words}",
      "example": "https://mcp.mdexa.com/v1/datasets?q=%E7%BB%93%E5%90%88%E4%BA%B2%E5%92%8C%E5%8A%9B"
    },
    "mcp": "https://mcp.mdexa.com/mcp",
    "anonymous_calls_per_minute": 20,
    "guide": "https://mdexa.com/zh/agents.md",
    "faq": "https://mdexa.com/zh/developers/faq/",
    "similar": [
      {
        "name": "CMtMedQA",
        "url": "https://mdexa.com/zh/datasets/cmtmedqa/",
        "json": "https://mdexa.com/zh/datasets/cmtmedqa.json"
      },
      {
        "name": "DISC-Med-SFT",
        "url": "https://mdexa.com/zh/datasets/disc-med-sft/",
        "json": "https://mdexa.com/zh/datasets/disc-med-sft.json"
      },
      {
        "name": "Huatuo-Llama-Med-Chinese 知识库问答与文献数据",
        "url": "https://mdexa.com/zh/datasets/huatuo-llama-med-chinese-data/",
        "json": "https://mdexa.com/zh/datasets/huatuo-llama-med-chinese-data.json"
      }
    ]
  }
}