Knowledge Resource Center for Ecological Environment in Arid Area
| 科学数据学术影响力评价体系研究 | |
| 其他题名 | The Research on Evaluation System of Scientific Data Academic Influence |
| 王毅萍 | |
| 出版年 | 2018 |
| 学位类型 | 硕士 |
| 导师 | 马建玲 |
| 学位授予单位 | 中国科学院大学 |
| 中文摘要 | 随着计算机、勘测等技术的进步，科学研究已经从计算科学向数据密集型知识发现的第四研究范式转变，科学数据已经成为推动学术研究发展与进步最为基础与重要的内容。在这样的背景下，学者们开展了大量关于科学数据共享、出版、引用、质量等方面的研究，为更广泛、更深入的利用科学数据创造有利条件。但目前存在科研人员数据共享积极性不高、引用意识薄弱等问题，严重制约了科学数据的共享与交流。除此之外，科研工作者对于高质量高水平科学数据的需求；资助机构为未来的资助资金流向进行评估判定的需求；数据存储与出版组织对于提高其组织影响力等的需求都激发出对科学数据影响力进行研究的必要性。为了进一步增强科研人员关于科学数据共享的激励、提高科学数据质量、为科学数据成果间的比较提供可能，本文综合运用定性定量相结合的方法，试图构建出一套科学、合理、实用、可操作的科学数据评价指体系。首先，笔者将本文的评价对象界定为科学数据集，同时对本文的研究框架进行构思，对全文研究方法与脉络有了较为清晰的把握。其次，通过大量文献调研，分析了国内外科学数据学术影响力评价现状，对现有的科学数据评价方法及工具进行总结。目前科学数据评价方法主要包括定性评价与定量评价，其中定性评价方法包括同行评议与环境评测方法，定量方法包括引文分析法、Altmetrics方法、数据仓储附加影响力计量方法、管理影响力计量方法；现有的科学数据评价工具可归纳为两大类，一种是以DCI等为代表的引文分析工具，另一种是以DLM、PlumX等为代表的Altmetrics工具。为下文评价指标体系的构建提供依据与指导。再次，结合科学数据的内涵、特征、相关关系与学术交流模型，分析科学数据在传播交流过程中的关键主体与影响因素，并从中提炼出关键评价指标，构建出包括数据原始影响力、数据使用影响力和第三方影响力在内的三级指标体系，其中一级指标3个，二级指标9个，三级指标22个。随后运用层次分析法，构造判断矩阵，完成专家问卷的设计、分发与回收，并利用yaahp软件计算出各项指标的权重，从而实现对科学数据学术影响力的定量化计算。随后，应用指标体系进行实证分析，实证分析领域为地球科学数据，样本数据为2010-2014年间我国寒区旱区科学数据中心、美国ORNL DAAC科学数据中心的科学数据集。应用本文所构建指标体系的同时结合地球科学数据特征，评价出各中心学术影响力Top10及综合学术影响力Top10的数据集，并对评价结果进行分析与验证。最后，结合前文的调研与所构建的指标体系，对当前制约我国科学数据学术影响力提升的因素进行分析，并据此提出提高我国科学数据学术影响力的建议。 |
| 英文摘要 | As the developing of computer and surveying technology, scientific researching has move forward to Fourth Research Paradigms of Data Intensive Knowledge Discovery from Computing Science, and Scientific Data already become foundation of academic research development. Against this backdrop, scholars do a lot of researches about scientific data sharing, publishing, citing, and quality, which improve the usage of scientific data widely and deeply. Nowadays, it is lacking enthusiasm of sharing and week reference consciousness that seriously restrict the sharing and exchange of scientific data. In addition, researchers’ need for high quality and high level scientific data, funding agencies’ need for evaluating the future funding flow, data storage and publishing organizations’ need for improving their influence, all these show the necessary to study the impact of scientific data. To enhance the researchers' incentive for scientific data sharing, improving the quality of scientific data, and providing the possibility for the comparison of scientific data results, this paper tries to construct a scientific, reasonable, practical and operable index system of scientific data evaluation, combining qualitative and quantitative methods. First, the author defines the evaluation object as scientific data set, and conceives the research framework, which helps form the clear methodology and thread of the full text.Secondly, over many literature research, this paper analyses the current situation of scientific data impact evaluation at home and abroad, and summarizes the existing methods and tools of it. At present, the evaluation methods of scientific data mainly contain qualitative and quantitative evaluation. Qualitative evaluation methods include peer review and environmental assessment methods, while Quantitative methods include Citation analysis, Altmetrics method, Data warehouse additional influence measurement and Management influence measurement. Current tools for evaluating scientific data can be classified into two categories, one is Citation analysis tools represented by DCI, and the other is Altmetrics tools represented by DLM and PlumX.Third, combining with the definition, the characteristics, the relationship and the academic exchange model of scientific data, this paper analyses the key subjects and influencing factors of scientific data, extracts key evaluation indexes, and finally construct three - level index system which includes the original influence of data, the influence of data use and third - party influence. Among them, the first level index has 3, the second-level index has 9, and the third-level index has 22. Then we use the Analytic Hierarchy Process to construct the judgment matrix, complete the design, distribution and recovery of the expert questionnaire, and use yaahp to calculate the weight of each index, so as to realize the quantitative calculation of the influence of scientific data. Forth, the index system is used for empirical analysis. The empirical data are Geo-scientific data, and the sample data are the scientific data sets between 2010 and 2014 from the Cold and Arid Regions Science Data Center at Lanzhou in China, and the Oak Ridge National Laboratory Distributed Active Archive Center （ORNL DAAC）in the United States. Based on the characteristics of the geoscience data, the Top10 and the Top10 data sets are selected and then the evaluation results will be analysed and validated.Finally, combined with the previous research and the index system constructed, this paper analyses the factors restricting the enhancement of scientific data in China, and puts forward some suggestions to improve the influence of scientific data in out country. |
| 中文关键词 | 科学数据 ; 学术影响力 ; 评价指标体系 |
| 英文关键词 | Scientific data Academic Influence Evaluation index system |
| 语种 | 中文 |
| 国家 | 中国 |
| 来源学科分类 | 图书馆学 |
| 来源机构 | 中国科学院文献情报中心 |
| 资源类型 | 学位论文 |
| 条目标识符 | http://119.78.100.177/qdio/handle/2XILL650/288045 |
| 推荐引用方式 GB/T 7714 | 王毅萍. 科学数据学术影响力评价体系研究[D]. 中国科学院大学,2018. |
| 条目包含的文件 | 条目无相关文件。 | |||||
| 个性服务 |
| 推荐该条目 |
| 保存到收藏夹 |
| 导出为Endnote文件 |
| 谷歌学术 |
| 谷歌学术中相似的文章 |
| [王毅萍]的文章 |
| 百度学术 |
| 百度学术中相似的文章 |
| [王毅萍]的文章 |
| 必应学术 |
| 必应学术中相似的文章 |
| [王毅萍]的文章 |
| 相关权益政策 |
| 暂无数据 |
| 收藏/分享 |
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。
