• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2006, Vol. 28 ›› Issue (9): 64-66.

• 论文 • 上一篇    下一篇

上下文无关文法推断中的几条启发规则及其应用

李志圣 陈永生   

  • 出版日期:2006-09-01 发布日期:2010-05-20

  • Online:2006-09-01 Published:2010-05-20

摘要:

本文先简要介绍了一种上下文无关文法的推断方法——逐步求精法,然后论述了递归概念在文法推断中的核心作用,并从递归概念的特殊性质出发提出了多条启发规则,能有效减少无效探求和与用户交互的次数,尤其适合于文法较复杂、例句集信息量较大的情况。这些启发规则同时也适用于对上下文无关文法的其它推断方法。

关键词: 文法推断 逐步求精 启发规则

Abstract:

This paper first introduces the stepwise refining method for inferring context-free grammars, and discusses the important role owned by the recursive  concept in the context-free grammars. Secondly, it presents several rules to improve the stepwise refining method in order to decrease the times of inqu iring of the user by the research on the specialty of the recursive concept. These rules are also suitable for other methods for inferring context-free  grammars.

Key words: (grammatical inference;stepwise refining, heuristic rules)