MNBVC-Wiki

这是本文档旧的修订版！

---

测试对象：open-cc和zhconv简繁体转换工具 测试字符：王力《古代汉语》附录简繁体对照字表 测试结果： （1）繁体-简体准确率：open-cc-0.8883；zhconv-0.8849 （2）简体-繁体准确率：open-cc-0.9695；zhconv-0.9814 繁体字以台湾繁体为准。

测试代码示例： from zhconv import convert

id_c_s2t = 0 id_c_t2s = 0 id = 0 with open(“characters.txt”, “r”,encoding='utf-8') as file:

  lines = file.readlines()
  for line in lines:
      id = id + 1
      char_s2t = convert(line.strip()[0],'zh-tw')
      if char_s2t == line.strip()[-1] or char_s2t == line.strip()[-2]:
          id_c_s2t = id_c_s2t + 1
      line2 = line[0:3]
      char_t2s = convert(line2.strip()[-1],'zh-cn')
      if char_t2s == line2.strip()[0]:
          id_c_t2s = id_c_t2s + 1
      # print(line.strip()[-2])

pre_s2t = id_c_s2t / id print(f'pre_s2t={pre_s2t}\n') pre_t2s = id_c_t2s / id print(f'pre_t2s={pre_t2s}\n')