Wondering if you guys ever considered incorporating any corpora info? For instance:
LIVAC Synchronous Corpus - esp. Chinese speech communities spread data, i.e.: which areas use what words.
Sketchengine - collocations, word combinations, synonyms, concordance features, etc.
LIVAC Synchronous Corpus - esp. Chinese speech communities spread data, i.e.: which areas use what words.
Sketchengine - collocations, word combinations, synonyms, concordance features, etc.