论文部分内容阅读
在复杂的数据环境中建立并维持非常高的数据质量是很昂贵并且往往不可能实现的。对质量的定量评估能够为制定改进措施的优先顺序提供重要的帮助。本文探索了一套方法,能够对数据质量的公正评估与效用导向评估两种方法进行评价。公正评估可以评价和衡量数据缺陷的程度,效用导向的评估能够衡量在一个特定的使用环境中,质量缺陷的存在对降低数据效用的程度。本文介绍的质量评估方法通过一个实际的校友数据库得到了实际验证。这个数据库是一个巨大的数据资源,它对校友关系和发起的承捐活动进行管理。质量评估的结果能够为这个数据库执行和管理改进数据质量的策略提供帮助。
Establishing and maintaining very high data quality in a complex data environment is expensive and often impossible. Quantitative assessments of quality can provide important help in prioritizing improvements. This article explores a set of methods that can evaluate both the fair evaluation of data quality and the utility-oriented evaluation. Just assessment can assess and measure the extent of data defects. Utility-oriented assessment can measure the extent to which the existence of quality defects reduces data utility in a particular environment of use. The quality assessment method described in this article is actually validated through an actual alumni database. This database is a tremendous data source that manages the alumni relationship and initiated donor activities. The results of the quality assessment can assist the database in implementing and managing strategies to improve data quality.