中文切分词典的最大匹配索引法

STUDY ON MAXIMUM MATCHING INDEX OF MACHINE LEXICON FOR CHINESE WORDS DISTINGUISHING

提高自然语言词典的检索速度是提高计算机处理系统速度的重要途径，考虑到汉语语词切分中使用词典的特点，参考“Trie索引”方法，提出了一种多层次的树型词典索引结构和最大匹配持查汉字串的检索方法.使用这种方法，切分系统每次都能查到所需，大大提高了切分的工作效率，在查一个登录的词时，也较一般的方法快.

A computer dictionary is the backbone for natural language processing sysems (NLPS). It is very important for NLPS to increase the searching speed of the dictionary. In view of the feature of the word distinguishing of Chinese sentences,and referring to the Trie index method,a multi-level-tree index structure and a relevant maximum-match index algorithm for Chinese word distinguishing are proposed. Using the method,the word distinguishing system can find what it needs at any time. Therefore it greatly increases the efficiency of Chinese word distinguishing systems,and it is also faster than other methods in any exisitng word inquiring.