正则文法在数据分析中的应用  

The application of regular grammar on data analysis

作　　者：谷长昱 刘建 

机构地区：[1]浙江天宇信息技术有限公司,浙江杭州310006

出　　处：《计算机时代》2015年第10期33-35,共3页Computer Era

摘　　要：以用词严谨、规范性极强的刑事判决书作为文本分析的对象,提取文书中的量刑情节语义。提出了一种异于依存句法分析的方法,该方法将句子抽象成若干元素,把这些元素的组成称之为句型结构。识别语义的过程就是识别结构类型,根据具体的结构类型分别提取其语义。该识别方法中,正则文法起到了关键作用,即采用正则表达式识别元素,使用正则文法定义结构类型。虽然该方法目前只应用在刑事判决书上,但也为类似问题的解决提供了思路。Taking the criminal judgment, which is carefully worded and strong normative, as a text analysis object, extracts the sentencing circumstances semantic from it. This paper proposes a method, in which a sentence is abstracted into several elements and the composition of the elements is called sentence structure, it is different from the dependency parsing. The process of recognizing semantics is to identify the structure type, according to the specific type of structure to extract the semantics respectively. In this recognition method, the regular grammar plays a key role that is to use regular expressions to identify elements and use regular grammar to define the structure type. Although this method is only used in the criminal judgment, it also provides a way to solve the similar problem.

关 键 词：数据分析 正则文法 正则表达式 句型 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]