数据库语义和词性双标注模型研究

The research of double tagging based on database semantic and part of speech

本文主要分为三个方面:语义词典的构建,词语标注的数据结构和数据库语义的标注与排歧算法。其中词典用来存储数据库的语义信息,通过程序调用以标注分词后的词语;词语标注的数据结构采用了动态的方式存储数据库语义,可节省内存空间并增强程序的可读性;对于数据库的歧义问题提出了一种利用相关词的语义确定歧义词的语义的方法,充分利用了词语之间的相互关系。

This article includes three aspects： the construction of the semantic dictionary; the data structure stored words and seman- tic; the method of tagging database semantic and avoiding ambiguity. The dictionary is used to store the semantic information and tag words have been segmented. The data structure uses dynamic structure to store database semantic in order to save the memory space and enhance the readability of the program. Put forward a method that uses related words to deal with ambiguity.