现代汉语缩略语自动识别的方法研究

Research methods about automatic identification of modern Chinese abbreviation

在中文信息处理领域,缩略语识别是研究中很重要的一个方面。针对缩略语词典资源稀少的现状,提出一种在生语料中自动抽取现代汉语缩略语的方法。首先获取候选缩略语的源短语候选集,然后利用基于上下文的源短语与缩略语配对方法,可自动生成一部缩略语词典,实验结果证明,该方法是一种相对"智能"的方法。

Identification of modem Chinese abbreviation is a very important study in Chinese information processing. Because of being short of abbreviation dictionary now, an approach is proposed, which would realize the automatic identification of modem Chinese ab- breviation. First abbreviation candidate is gained, then the abbreviation dictionary is automatically produced based on the context. The experiment show the approach is a relatively ＂smart＂ approach.