| 近似重复记录的增量式识别算法 |
| |
| 引用本文： | 许向阳,佘春红. 近似重复记录的增量式识别算法[J]. 计算机工程与应用, 2003, 39(12): 191-193,220 |
| |
| 作者姓名： | 许向阳 佘春红 |
| |
| 作者单位： | 华中科技大学计算机学院数据库与多媒体技术研究所,武汉,430074 |
| |
| 基金项目： | 国家科技攻关计划项目“科技部科技电子政务系统关键技术及应用系统的研究”(编号:2001BA110B01) |
| |
| 摘 要： | 摘要数据清理是数据仓库中的一个重要研究内容,近似重复记录的识别是其中的一个技术难点。文章介绍了近邻排序方法,并以此为基础,研究了在数据模式与匹配规则不变的前提下,数据源动态增加时近似重复记录识别问题,提出了一种增量式算法IMPN(IncrementalMulti-Passsorted-Neighborhood)。文章最后给出了实验结果。
|
| 关 键 词： | 数据清理 近似重复记录 增量式识别 特征记录 |
| 文章编号： | 1002-8331-(2003)12-0191-03 |
Incremental Algorithm for Detecting Approximately Duplicate Database Records |
| |
| Xu Xiangyang She Chunhong. Incremental Algorithm for Detecting Approximately Duplicate Database Records[J]. Computer Engineering and Applications, 2003, 39(12): 191-193,220 |
| |
| Authors: | Xu Xiangyang She Chunhong |
| |
| Abstract: | Data cleaning is an important area of data warehouse.Detecting approximately duplicate database records is one of technology difficulties.This paper introduces sorted-neighborhood method.Based on this idea,it studies the prob-lem for detecting approximately duplicate records while receiving increments of data with no changes in data schema and matching rule-set,and presents an incremental algorithm for detecting the records.Finally,it gives out the experi-mental results. |
| |
| Keywords: | Data cleaning Approximately duplicate records Incremental detection Representative record |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
