Information Evaluation Based on Sources Dependence

Information Evaluation Based on Sources Dependence
复制标题

DOI:
10.3724/sp.j.1016.2012.02392
复制
发表时间:
2012
期刊:
Chinese Journal of Computers
影响因子:
--
通讯作者:
Zhiqiang Zhang;Lixia Liu;Xiaoqin Xie;Haiwei Pan;Yixiang Fang
Zhiqiang Zhang;Lixia Liu;Xiaoqin Xie;Haiwei Pan;Yixiang Fang
中科院分区:
其他
文献类型:
--
作者:
Zhiqiang Zhang;Lixia Liu;Xiaoqin Xie;Haiwei Pan;Yixiang Fang

文献摘要

被引文献

相似文献

许多数据管理应用需要集成来自多个数据源的数据。每个数据源都提供了一组值,而不同的源往往会提供冲突的值。为了将高质量的数据呈现给用户,数据集成系统能够解决冲突并发现真值是至关重要的。本文对现有算法进行了改进,使用了一种新的投票算法,并考虑了相同值的不同表达方式,如人名。实验结果表明,该算法对于发现冲突值中的真值是非常有效的。
Many data management applications require integrating data from multiple sources.Each of these sources provides a set of values and different sources can often provide conflicting values.To present quality data to users,it is critical that data integration systems can resolve conflicts and discover true values.In this paper,we improve the existing algorithm with using a new voting algorithm and considering the diverse expressions of the same value,e.g.person's name.The experiment results shown it is very effective for discovering true values among conflicting values.