一种集成WordNet到受控自然语言系统的算法

A system integrating WordNet into controlled natural language

首先分析英语单词词性的分类和特性,定义出受控自然语言的词库模型;然后结合WordNet词库特性提出基于WordNet的受控自然语言词库构建方法;最后提出结合基于词库的上下文无关文法,并通过其实现词库解析算法.试验结果表明,按照本算法集成WordNet词库到ACE受控自然语言系统,能显著地提高受控自然语言的识别率,同时降低受控自然语言词库的构建和维护门槛,具有可行性和实用性.所提算法具有通用性和较强的可移植性,可以很容易移植到其他受控自然语言系统.

irstly controlled natural language （CNL） lexicon mode is defined after analyzing the English words part of speech classification and characteristics.Then the way to build a CNL lexicon based on WordNet is proposed by utilizing WordNet features;At last,by lexicon-based context-free grammar （CFG） proposed by this paper,an algorithm to parse lexicon is implemented.WordNet was integrated into the attempto controlled english （ACE） in accordance with the algorithm proposed.Experimental results show that this algorithm can significantly improve the recognition rate of CNL,while reducing the CNL lexicon construction and maintenance of the threshold,with feasibility and practicality.