摘要
从排序属性的选择、匹配方法、相似度计算、检测和处理相似重复记录以及实验结果几个方面,阐述了一种有效检测汉语相似重复记录的方法。
This paper expounds an efficient approach for detecting approximately duplicate Chinese records from aspects of the selection of sort properties,matching methods,similarity computation,detection and treatment of approximately duplicate records,and test results,etc.
出处
《科技情报开发与经济》
2011年第21期133-135,共3页
Sci-Tech Information Development & Economy
关键词
相似重复记录
匹配
排序属性
approximately duplicate records
matching
sort properties
