一种基于编码规则的中文地址清洗方法

由于中文地址命名的不规范性和中文的书写特点,造成中文地址的清洗工作异常困难.中文地址是由地址元素和特征字两部分构成的,在对中文地址预处理的基础上,通过制定中文地址字符编码规则,提出对中文地址字符进行编码,在地址元素后添加特征字代码,利用编码规则对地址代码进行清洗,最后根据编码结果对代码进行译码,达到清洗的目的.利用某常住人口地址进行验证,实验结果证明清洗效果良好.

A coding rule-based cleaning approach to Chinese address

Because of the non - standard Chinese address and the writing features of Chinese, Chinese ad- dress cleaning is rather difficult. Chinese address consists of the address element and signature words. Through address character encoding rules, the encoding for address characters which is on the basis of Chi- nese address pre -processing is proposed. Clean the address code by using rule in which signature words code is added after address element. Finally according to the encoding result, decode the code to achieve the purpose of cleaning. The experiment which uses addresses of some resident proves that it has good cleaning effect.