科学数据众包处理研究

发布时间：2018-02-01 21:15

本文关键词： 众包科学大数据数据处理人才筛选质量评估　出处：《计算机研究与发展》2017年02期 　论文类型：期刊论文

【摘要】：获取科学数据的最终目的是根据具体需要从数据中提取有用的知识,并将这些知识应用到具体的领域中,帮助决策制定者制定决策.由于科学数据规模越来越大,而且呈现结构复杂的特点,如半结构化或非结构化,难以通过计算机实现自动化处理.众包通过高效调用人力资源,成为进行科学大数据众包处理的解决方案之一.针对科学大数据众包处理的特点,围绕人才筛选机制、任务处理模式和结果评估策略3方面对科学数据众包体系进行研究,并通过地理空间数据云平台开展地学领域的基于众包的遥感影像信息提取实验.研究表明,科学数据不仅能够通过众包模式来进行处理,而且通过合理的设计众包流程能够获得高质量的数据结果.
[Abstract]:The ultimate goal of obtaining scientific data is to extract useful knowledge from the data according to specific needs and to apply that knowledge to specific areas. Help decision makers to make decisions. Because of the growing scale of scientific data and complex structural characteristics such as semi-structured or unstructured. It is difficult to realize automatic processing by computer. Crowdsourcing through the efficient transfer of human resources has become one of the solutions for scientific big data crowdsourcing processing. Aiming at the characteristics of scientific big data crowdsourcing processing. The scientific data crowdsourcing system is studied in three aspects: talent selection mechanism, task processing model and result evaluation strategy. And through the geospatial data cloud platform to carry out the geoscience field based on crowdsourcing remote sensing image information extraction experiment. The research shows that scientific data can not only be processed by crowdsourcing mode. Furthermore, high-quality data results can be obtained through reasonable design of crowdsourcing process.
【作者单位】：中国科学院计算机网络信息中心;中国科学院大学;天津大学管理与经济学部;
【基金】：国家重点研发计划项目(2016YFB1000600,2016YFB0501900) 国家自然科学基金项目(71571133) 中国科学院战略性先导科技专项(XDA06010307)~~
【分类号】：TP751;TP311.13
【正文快照】： This work was supported by the National Key Research Program of China(2016YFB1000600,2016YFB0501900),the NationalNatural Science Foundation of China(71571133),and the Strategic Priority Research Program of the Chinese Academy of Sciences(XDA06010307).(zj

【相似文献】