基于模板知识的带噪音半结构文本数据自动分词方法  

A method of segment-parsing for SEMI-structured data with random noise based on knowledge about template

作　　者：解辉[1] 王晓英[1] 金鑫[1] 

机构地区：[1]青海大学计算机技术与应用系,青海西宁810016

出　　处：《微型机与应用》2015年第17期89-91,95,共4页Microcomputer & Its Applications

基　　金：国家自然科学基金项目(61363019);青海省创新能力促进计划项目(2014-ZJ-718;2014-ZJ-941Q)

摘　　要：随机噪音会引起半结构化数据结构和语义的变化,从而造成半结构化数据提取或向结构化自动转换障碍,为此本文在分析噪声对半结构化文本数据带来的影响基础下,提出了一种基于语法模板结构知识的自动化分词处理思路和方法 ,可以提高带噪音的半结构化文本数据自动分词准确性,可为类似问题提供参考。Random noise always bring challenges to segment auto parsing for SEMI-structured data, and cause hindrance for extracting SEMI-structured data and autochange to structured data. With analyzing the effects bring by noise to the SEMI-structured data, this paper introduced a new method based on knowledge about the template that the SEMI-structured data organized by. The method could handle a set of noise and raise the accuracy rate for segment-parsing of SEMI-structured data, so it could be helpful for the research about homologous problems.

关 键 词：半结构数据 分词 模板 噪音 

分 类 号：TP317[自动化与计算机技术—计算机软件与理论]