{
  "title": "字有来路——常用汉字的构形与生命史",
  "subtitle": "常用汉字的构形与生命史",
  "sampleSize": 123,
  "componentCount": 18,
  "edgeCount": 73,
  "corpus": {
    "uniqueCharacters": 9933,
    "totalCharacters": 193504018,
    "top100Cumulative": 41.770331094624,
    "top250Cumulative": 60.58768557457
  },
  "sample": {
    "medianFrequencyRank": 577,
    "outsideTop1000": 32,
    "withStrokeData": 123
  },
  "updated": "2026-10-18",
  "sources": [
    {
      "name": "Jun Da 现代汉语字频表",
      "detail": "Modern Chinese Character Frequency List，语料 193,504,018 字，9,933 个不同汉字，数据更新于 2004-03-30。",
      "url": "https://lingua.mtsu.edu/chinese-computing/statistics/char/list.php?Which=MO"
    },
    {
      "name": "Make Me a Hanzi · dictionary.txt",
      "detail": "由 Unihan 与 CJKlib 派生的构形、部首、拼音与字源提示数据；LGPL v3。",
      "url": "https://github.com/skishore/makemeahanzi"
    },
    {
      "name": "Make Me a Hanzi · graphics.txt",
      "detail": "由 Arphic PL KaitiM GB / UKai 派生的笔顺矢量数据；Arphic Public License。",
      "url": "https://github.com/skishore/makemeahanzi"
    },
    {
      "name": "Unicode Unihan Database",
      "detail": "汉字读音、部首与字典信息的上游来源之一。",
      "url": "https://www.unicode.org/charts/unihan.html"
    }
  ],
  "cautions": [
    "本作选取 123 个常用字作为策展样本，不是完整字典，也不以字频排名判断文化价值。",
    "字频来自 2004 年语料快照；同一字在不同语体、时代与地区中的频率会变化。",
    "构形数据采用 IDS 表意文字描述序列；部分生僻部件在上游数据中以“？”标记，本作不擅自补全。",
    "“字族”指共享字形部件的关系，不等同于严格的历史词源关系。",
    "甲骨文意象图为依据常见字形特征的概念重绘，不是考古拓本；书体年表用于展示演变脉络，不作为文字学定论。"
  ]
}