CIF21 DIBBs: An Infrastructure for Computer Aided Discovery in Geoscience

下一代地球科学需要处理来自地基和天基传感器网络的快速增长的数据量。随着现实世界的现象被映射到数据，科学发现过程本质上变成了一个跨多维数据集的搜索过程。因此，从这片数据海洋中提取有意义的发现需要高效和可扩展的机器辅助，以增强人类对上下文的理解。这对于测试新的假设以及检测新事件和监测自然灾害都是必要的。该项目开发了一种计算机辅助发现方法，为科学家提供更好的支持来回答以下问题：从已识别的特征可以得出什么样的推论？ 一个发现意味着什么？它如何融入大的理论图景？它是否与以前建立的模型和调查结果相矛盾或相证实？如何有效地测试概念和想法？为了实现这一点，科学家可以编程方式表达假设的地球科学场景，约束条件和模型变化。这种方法有助于委托自动探索的组合搜索空间的可能的解释并行各种数据集。此外，可编程爬虫可以在基于云的基础设施上扩展对有趣现象的搜索和发现。计算机辅助发现原型在地球空间科学的案例研究中进行了评估，包括使用组合GPS，光学和地球空间雷达数据探索空间和时间结构。

Next-generation Geoscience needs to handle rapidly growing data volumes from ground-based and space-based sensor networks. As real-world phenomena are mapped to data, the scientific discovery process essentially becomes a search process across multidimensional data sets. The extraction of meaningful discoveries from this sea of data therefore requires highly efficient and scalable machine assistance to enhance human contextual understanding. This is necessary both for testing new hypotheses as well as for the detection of novel events and monitoring for natural hazards.This project develops a computer-aided discovery approach that provides scientists with better support to answer questions such as: What inferences can be drawn from an identified feature? What does a finding mean and how does it fit into the big theoretical picture? Does it contradict or confirm previously established models and findings? How can concepts and ideas be tested effectively? To achieve this, scientists can programmatically express hypothesized Geoscience scenarios, constraints, and model variations. This approach helps delegate the automatic exploration of the combinatorial search space of possible explanations in parallel on a variety of data sets. Furthermore, programmable crawlers can scale the search and discovery of interesting phenomena on cloud-based infrastructures. The computer-aided discovery prototype is evaluated in case studies from Geospace science, including the exploration of structures in space and time using combined GPS, optics, and Geospace radar data.