The Lancaster Mandarin Chinese Corpus

简介：
《The Lancaster Mandarin Chinese Corpus》是由兰卡斯特大学创建的汉语语料库，收录了大量现代标准普通话的真实语言材料，为中文研究与教学提供了宝贵资源。 The Lancaster Corpus of Mandarin Chinese (LCMC) is designed to serve as an equivalent resource for modern Mandarin Chinese, similar to the FLOB and FROWN corpora which are used for British and American English. This corpus can be utilized in monolingual research on contemporary Mandarin Chinese or in cross-linguistic studies comparing Chinese with British/American English. The LCMC includes a sample of 15 categories of written texts, such as news articles, literary works, academic essays, and official documents published in the Peoples Republic of China during the early 1990s. The total size is approximately one million words. It employs the same sampling criteria and time frame used by FLOB/FROWN. The corpus includes markup for text categories, sample file numbers, paragraphs, sentences, and tokens. Linguistic annotations cover tokenization and part-of-speech tagging at the word level with orthographic and morphological details. The Institute of Computing Science Chinese Lexical Analysis System (ICTCLAS) from the Chinese Academy of Sciences provided the tagging system. The corpus is encoded in Unicode (UTF-8) and marked up using XML. It comes with a User Manual that explains design specifications and part-of-speech tags. Xairas parser was used to validate the XML structure, while all aspect markers were manually verified for accuracy in their part-of-speech tagging.