HKSeg

It's a collaborative ongoing work.
† Mainly contributed by Minghao
‡ Mainly contributed by Randy

Dictionary

• dict_celebrity †
  • Hong Kong celebrity
• dict_institution †
• dict_location_hk †
  • Includes locations related to Hong Kong. They don't have to be in Hong Kong physically.
• dict_politics ‡
• dict_covid19_vaccine †
• dict_covid19 †
• dict_general_en ‡ †
• dict_general_zh ‡ †

Stopword

• ad_words †
  • Summaried keywords related to advertisement and promotional contents appears frequently in Hong Kong public media.
  • Keywords are in regular expression format.
• stopwords_canto ‡ †
  • Stopwords in Cantonese.
• stopwords_en ‡ †
  • Stopwords in English.
• stopwords_simple †
  • A simplified list of Cantonese stopwords. Numbers are included.

jieba tools

• jieba_formatter.py †
  • A script made for solving a mistake when loading user-defined dictionary with jieba (jieba3k 0.35.1).