Summary: ZHONG is a mid-sized grammar, with targeted coverage. Much of its development has been centered around educational data, which means it is able to provide adequate analyses for most of the sentence patterns found in early to intermediate educational materials. From a linguistic point of view, ZHONG offers in-depth analyses of interesting and well-studied linguistic phenomena including: serial-verbs, 的 (DE) constructions, 把 (BA) constructions, interactions between aspect and negation, sentence final particles, and interrogatives.
Input: ZHONG expects its input to be segmented at the word level. This means that words must be separated by spaces (see some of the examples below). Some of our analyses expect input to be tokenized differently than most Mandarin Chinese word segmentation tools (e.g., 这个 should be tokenized as two tokens 这 and 个). To the best of its ability, the grammar uses REPP (internal preprocessing) to help further segment the input (e.g., if 这个 is provided as a single token, it will be internally split as two tokens).
Examples:
- 我 正在 看 电视 呢 。
- 这 个 商场 很 大 。
- 我们 还 要 三 碗 汤 。
- 你 会 不 会 开车 ？
- 语言 学院 有 多少 个 系 ？