SGER: Metadata-Driven Approach to Discovery-Oriented Exploration of Massive Data Sets

SGER：元数据驱动的海量数据集发现导向探索方法

• 批准号: 0841031
• 负责人: Ruth West
• 金额: --
• 依托单位: University of California-Los Angeles
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2008
• 资助国家: 美国
• 起止时间: 2008-09-01 至 2010-02-28
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0841031&HistoricalAwards=false
• 关键词: SGER; Metadata; Driven; Approach; Discovery

Multi-scale, multi-modal, multi-site science, such as terrestrial observatories and other dynamic and adaptive sensor-based application, high resolution sub-cellular imaging, or genomic analyses of communities of organisms, builds large data sets that can serve as the basis for in silico exploration and analysis. Often the data collections represent a wide range of disciplines, yet there are often a limited number of interpretations, guided by an individual scientist?s expertise. Intuitive exploration that transcends disciplinary boundaries and expertise could enhance the process of interdisciplinary collaboration, as well as drive the process of discovery in new directions. This proposal seeks to develop a dynamic metadata-based approach for intuitive, interactive, and immersive multi-scale, multi-modal data exploration application to a wide range of data sets and their associated metadata. The original focus will be on metagenomics data from the Global Ocean Survey. The development of the environment will enable a different dynamic sorting and sifting of data rather than relying solely on known or expected features. Such open-ended exploration occurs not only in observational science, but also in tasks ranging from network intrusion detection. Students will be engaged in the research and collaboration, with dissemination in public cultural activities, as well as through more traditional academic venues.

多尺度、多模式、多地点的科学，如地面观测站和其他基于动态和自适应传感器的应用、高分辨率亚细胞成像或生物群落的基因组分析，建立了大型数据集，可以作为计算机探索和分析的基础。通常，数据收集代表了广泛的学科，然而，在单个科学家的指导下，往往有有限数量的解释。专业知识。超越学科界限和专业知识的直觉探索可以促进跨学科合作的进程，并推动新方向的发现过程。本提案旨在开发一种动态的基于元数据的方法，用于直观、交互和沉浸式的多尺度、多模态数据探索应用于广泛的数据集及其相关元数据。最初的重点将是来自全球海洋调查的宏基因组学数据。环境的发展将使数据的动态排序和筛选成为可能，而不是仅仅依赖于已知或预期的特征。这种开放式探索不仅出现在观测科学中，也出现在网络入侵检测等任务中。学生将参与研究和合作，在公共文化活动中传播，以及通过更传统的学术场所传播。