A Fast algorithm for detection of Chinese lihe words

摘要

In the Chinese linguistic study, there is a kind of words that are called lihe words. These words can be separated by certain modifier words such as adjective, quantifier, and so on, but the meaning of the word is not changed. If we separated the part of the lihe Chinese word as an independent word, it is not reasonable and easily makes some unnecessary ambiguous segmentation. So how to detect lihe word quickly and effectively is a real problem in Chinese natural language processing. In this paper, we introduced a model of processing lihe words, which use Semantics Extracting Tree to detect the lihe words quickly and accurately.