Chinese Lettered-words Extraction for Language Monitoring

摘要

Lettered-words are frequently used in Chinese.Lettered-word falls into two parts.They are lettered-word with Chinese characters and letteredword without Chinese characters.Chinese characters in lettered-word have no specialization.When letteredword with Chinese characters are scattered in Chinese texts,it is difficult to recognize the boundaries.As a result,lettered-word with Chinese characters becomes a difficulty for lettered-words identification and extraction.In this paper,a method to extract lettered-words with Chinese characters and lettered-words without Chinese characters separately is proposed for the first time.An experiment on language monitoring of lettered-words using shows that the proposed method achieves a high recall and precision.