Python 偵測中文字、姓名、拼音教學與範例

介紹如何使用 Python 偵測文字中的中文姓名或拼音，用於個人資料之去識別化。

檢查中文拼音

使用 Python 的 re.finditer 以正規表示法將資料中的每一個單字切出來，再逐一檢查是否為中文字或是拼音，而檢查中文拼音所使用到的對照表，可以從外交部領事事務局的護照外文姓名拼音對照表取得，亦可直接下載我整理好的 CSV 檔案。

import re
import csv

# 拼音對應表
PinYinMap = {}

# 開啟拼音對照表 CSV 檔案
with open('pinyin-utf8.csv', newline='') as csvfile:

    # 讀取 CSV 檔內容，將每一列轉成一個 dictionary
    rows = csv.DictReader(csvfile)

    # 以迴圈輸出指定欄位
    for row in rows:
        # 國音第一式,漢語拼音,通用拼音,國音第二式,WG
        if row['漢語拼音'] in PinYinMap:
            PinYinMap[row['漢語拼音']].add(row['國音第一式'])
        else:
            PinYinMap[row['漢語拼音']] = {row['國音第一式'],}

        if row['通用拼音'] in PinYinMap:
            PinYinMap[row['通用拼音']].add(row['國音第一式'])
        else:
            PinYinMap[row['通用拼音']] = {row['國音第一式'],}

        if row['國音第二式'] in PinYinMap:
            PinYinMap[row['國音第二式']].add(row['國音第一式'])
        else:
            PinYinMap[row['國音第二式']] = {row['國音第一式'],}

        if row['WG'] in PinYinMap:
            PinYinMap[row['WG']].add(row['國音第一式'])
        else:
            PinYinMap[row['WG']] = {row['國音第一式'],}

# 測試資料（王曉明）
data = "My name is Xiao-Ming Wang."

# 以正規表示法取出每一個單字
for x in re.finditer(r'\b\w+\b', data):
    # 取得單字
    word = x.group()

    # 檢查單字是否為中文拼音
    if word.upper() in PinYinMap:
        # 顯示拼音對應的注音
        print(f"找到中文拼音：{word} {PinYinMap[word.upper()]}")

找到中文拼音：Xiao {'ㄒㄧㄠ'}
找到中文拼音：Ming {'ㄇㄧㄥ'}
找到中文拼音：Wang {'ㄨㄤ'}

檢查中文字

在 Unicode 編碼中漢字的範圍是 U+4E00 到 U+9FFF，我們可以透過以下正規表示法來檢查文字資料是否包含漢字：

import re

# 測試資料
data = "您好，我是王曉明。"

# 以正規表示法取出每一個中文字
for x in re.finditer(r'[\u4E00-\u9FFF]', data):
    # 取得中文字
    word = x.group()
    print(f"找到中文字：{word}")

找到中文字：您
找到中文字：好
找到中文字：我
找到中文字：是
找到中文字：王
找到中文字：曉
找到中文字：明

以下是另外一種寫法，可以不需要使用到正規表示法（regular expression）：

# 測試資料
data = "您好，我是王曉明。"

for c in data:
    # 檢查是否為中文字
    if u'\u4E00' <= c <= u'\u9FFF':
        print(f"找到中文字：{c}")

找到中文字：您
找到中文字：好
找到中文字：我
找到中文字：是
找到中文字：王
找到中文字：曉
找到中文字：明

姓氏檢查

百家姓中收錄了幾百個常見的姓氏，可用於檢驗姓名，而近代的姓氏則可參考台灣姓氏的統計。我們可以透過姓氏的檢查，輔助判斷資料中是否含有姓名：

# 台灣常見姓氏
LastName = [
    "陳","林","黃","張","李","王","吳","劉","蔡","楊","許","鄭",
    "謝","洪","郭","邱","曾","廖","賴","徐","周","葉","蘇","莊",
    "呂","江","何","蕭","羅","高","潘","簡","朱","鍾","游","彭",
    "詹","胡","施","沈","余","盧","梁","趙","顏","柯","翁","魏",
    "孫","戴","范","方","宋","鄧","杜","傅","侯","曹","薛","丁",
    "卓","阮","馬","董","温","唐","藍","石","蔣","古","紀","姚",
    "連","馮","歐","程","湯","黄","田","康","姜","白","汪","鄒",
    "尤","巫","鐘","黎","涂","龔","嚴","韓","袁","金","童","陸",
    "夏","柳","凃","邵","錢","伍","倪","溫","于","譚","駱","熊",
    "任","甘","秦","顧","毛","章","史","官","萬","俞","雷","粘",
    "饒","張簡","闕","凌","崔","尹","孔","歐陽","辛","武","辜",
    "陶","易","段","龍","韋","葛","池","孟","褚","殷","麥","賀",
    "賈","莫","文","管","關","向","包","丘","梅","范姜","華","利",
    "裴","樊","房","全","佘","左","花","魯","安","鮑","郝","穆",
    "塗","邢","蒲","成","谷","常","閻","練","盛","鄔","耿","聶",
    "符","申","祝","繆","陽","解","曲","岳","齊","籃","應","單",
    "舒","畢","喬","龎","翟","牛","鄞","留","季","覃","卜","項",
    "凃","喻","商","滕","焦","車","買","虞","苗","戚","牟","雲",
    "巴","力","艾","樂","臧","司","樓","費","屈","宗","幸","衛",
    "尚","靳","祁","諶","桂","沙","欒","宮","路","刁","時","龐","瞿"]

# 測試資料
data = "您好，我是王曉明。"

# 對每個姓氏逐一檢查
for ln in LastName:
    if ln in data:
        print(f"含有姓氏：{ln}")

含有姓氏：王

姓名拼音對照表

以下是外交部領事事務局的護照外文姓名拼音對照表，有注音與各種拼音的對照，亦可直接下載我整理好的 CSV 檔案。

參考資料

• The Will Will Web：整理 Unicode 經常會使用到的內碼區域並透過 Regex 自動比對文字