基于双哈希结构的整词二分词典机制  

Binary-seek-by-word dictionary mechanism based on double-Hash

作　　者：张贤坤[1] 李亚南[1] 田雪[1] 

机构地区：[1]天津科技大学计算机科学与信息工程学院,天津300222

出　　处：《计算机工程与设计》2014年第11期3956-3960,共5页Computer Engineering and Design

基　　金：天津市科技型中小企业创新资金基金项目(12ZXCXGX33500)

摘　　要：为提高逆向最大匹配分词算法的速率,分析传统的整词二分词典机制,针对哈希结构尾字词典的不足,提出一种基于双哈希结构的整词二分词典机制,并改进对应的逆向最大匹配算法。利用记录词长的尾字词典机制,使用双哈希结构保存词条,降低词典的词条匹配时间复杂度。实验结果表明,该词典机制将逆向最大匹配算法的分词速率提高近一倍。To improve the rate of the reverse maximum matching algorithm, for the insufficiency of the last word dictionary based on the Hash structure, a binary-seek-by-word dictionary mechanism for the word segmentation based on double-Hash was proposed by analyzing traditional dictionary mechanisms, and the corresponding reverse maximum matching algorithm was improved. The double-Hash structure was used to save entries By making advantage of the last word dictionary mechanism that recorded words＇ length, the time complexity of string finding was reduced. Experimental results show that the mechanism nearly doubles the rate of the reverse maximum matching algorithm.

关 键 词：中文分词 词典机制 尾字词典 哈希结构 逆向最大匹配算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]