数据与计算发展前沿 ›› 2021, Vol. 3 ›› Issue (2): 68-76.

doi: 10.11871/jfdc.issn.2096-742X.2021.02.008

• 管理决策与智能应用专刊 • 上一篇    下一篇

ARP数据治理体系研究与实践

孙健英()   

  1. 中国科学院计算机网络信息中心,北京 100190
  • 收稿日期:2021-03-01 出版日期:2021-04-20 发布日期:2021-05-18
  • 通讯作者: 孙健英
  • 作者简介:孙健英,中国科学院计算机网络信息中心,高级工程师,硕士研究生导师,管理信息化部副主任,主要研究方向为智慧管理、软件工程,当前主要从事新一代ARP管理业务研究。
    独立编写本文。
    SUN Jianying is currently the senior engineer, master super-visor, and the deputy Director of Management informatization Deparment, Computer Network Information Center, Chinese Academy of Sciences. Her research interests cover intelligent management and software engineering.
    She accomplished the paper independently.
    E-mail: jysun@cnic.cn

Research and Practice on ARP Data Governance System

SUN Jianying()   

  1. Computer Network Information Center, Chinese Academy of Sciences, Beijing 100190, China
  • Received:2021-03-01 Online:2021-04-20 Published:2021-05-18
  • Contact: SUN Jianying

RichHTML

18

PDF

200

摘要：

【目的】信息化项目的显著成果是形成了大量数据资源,提高数据质量,充分挖掘数据资源,辅助管理决策,推动数据赋能是信息化纵深发展的目标。【背景】ARP（Academia Resource Planning）系统是中国科学院院属单位日常科研管理的工作平台,系统中存储了十余年的管理数据,但由于数据质量问题制约了数据分析的精准性,如何提升数据质量已成为应用的研究热点。【方法】本文研究管理信息化应用的数据质量管理标准及评估流程,以ISO/IEC25024数据质量模型为依据,结合中国科学院科研管理信息化特点,利用灰色关联度和相关系数建立指标排序模型,实现指标关联分析和主因子判定,并按照AHP层次分析法得出系统数据质量评价模型。【结果】依据本文所提出的数据质量评估模型,对ARP系统数据质量进行测评并进行可视化展示。结果显示,治理后数据基本实现智能决策辅助支持。数据质量在完整性方面得到大幅度提升,但在数据正确性方面还需改善。【结论】构建场景化、系统性的评估指标体系,可有效促进数据质量的提升,并以从用户视角建立的评估模型使数据治理定量评估更加科学。

关键词: ARP, 数据治理体系, 数据质量, 指标关联分析, 评估方法

Abstract:

[Objective] The most valuable result of the informatization project is the massive data resources accumulated. Improving the quality of data, exploring research data resources, assisting management decisions and promoting data empowerment are the purpose of further development in informatization. [Context] ARP (Academia Resource Planning) is the working platform for daily scientific research management of units affiliated to the Chinese Academy of Sciences. More than ten years of management data is stored in the system but the quality of data restricts the accuracy of data analysis. Thus, how to improve data quality has become a hot topic. [Methods] Based on the ISO/IEC25024 data quality model, this paper focuses on data quality management standards and evaluation processes for information management applications. Combined with the characteristics of scientific research management informatization of the Chinese Academy of Sciences, this paper established an index ranking model by using grey relational degree and correlation coefficient. [Results] The results of data quality measurement and visualization of ARP system based on the proposed data quality assessment model show that the data after governance can basically satisfy the needs of intelligent decision support. Data quality has been greatly improved in terms of completeness, but more improvement is required in terms of data accuracy. [Conclusions] A scenario-based and systematic evaluation index system can improve data quality. Besides, the quantitative assessment of data governance could be more rational and effective if the evaluation model is built from the perspective of users.

Key words: ARP, data governance system, data quality, index correlation analysis, evaluation method