基于模糊匹配的数据清洗

Data cleaning based on fuzzy match

提出了一种新的模糊匹配相似度函数及一种高效的模糊匹配算法 ,该算法能够高效的寻找与输入元组最相似的L个参考元组 ;为了提高查询效率 ,提出了对参照关系建立错误容错索引。

This paper proposes a new fuzzy match similarity function and an efficient algorithm which can quickly look up L tuples in reference tables.In order to improve querying efficiently,this paper creates the error tolerant index basing reference relations.