CoPe EAGER: Collaborative Research: COMET: the Coastlines and people Open data and MachinE learning sprinT

CoPe EAGER：协作研究：COMET：海岸线和人类开放数据和机器学习冲刺

基本信息

批准号：
1939954
负责人：
Evan Goldstein
金额：
$ 17.31万
依托单位：
University of North Carolina Greensboro
依托单位国家：
美国
项目类别：
Standard Grant
财政年份：
2019
资助国家：
美国
起止时间：
2019-09-15 至 2023-08-31
项目状态：
已结题

来源：
https://www.nsf.gov/awardsearch/showAward?AWD_ID=1939954&HistoricalAwards=false
关键词：
CoPe EAGER Collaborative Research COMET

项目摘要

Timely release of research results is important to advance understanding of the impacts of climate change and sea level rise on coastlines and the communities that live there. The growing library of open, freely accessible data and analysis tools (i.e., code) enables scientists to investigate a range of societally relevant questions at the intersection of Coastlines and People. This project pilots an incubator approach to catalyzing data-driven research and creating networks of researchers ready to tackle the complex problems of the coast. The program is modeled on other 'science sprints', where teams of researchers assemble to transform an idea into open, freely accessible research products within a short, fixed time window - thereby accelerating scientific advances. This project will advance science in three ways: 1) By creating cohorts of scientists using data-driven approaches to address the interdisciplinary problems along the coast; 2) Scientists at each event will create open tools, code, deliverables and data products, creating freely available methods and knowledge; 3) Multiple events and iteration between events will enable evaluation of the sprint approach, and its success in producing science at the intersection of Coastlines and People.The three sprint events are focused on quick turn-around research using open data and machine learning, and will take advantage of the vast data volumes available through data.gov and other FAIR (Findable, Accessible, Interoperable, Re-usable) sources. The objective is to produce results and deliverables rapidly. Applications from the scientific community will be solicited for each of the three planned events, and selection of cohorts will prioritize having representation of a diverse set of fields and perspectives. Each event will adhere to a Code of Conduct that will additionally include an 'open by default' statement for code, data, and reports generated at the sprint. At each event, participants will break into small groups to spend 72 hours working on selected projects. Groups will produce oral and written reports, as well as associated open source code at the end of each event. Outcomes from each event will be measured using surveys (pre- and post- event), and by following the use of digital object identifiers associated with the open deliverables from each event. Datasets used by the participants will also be collected and curated on a publicly available website as a crowd-sourced list of relevant open data. The three sprint events will take place in North Carolina and Colorado. A range of external collaborators will interact and network with participants.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

及时释放研究结果对于促进对气候变化和海平面上升对海岸线和居住在那里的社区的影响的理解非常重要。开放，自由访问的数据和分析工具（即代码）的日益增长的库使科学家能够在海岸线和人的交集中调查一系列社会相关问题。该项目飞行员是一种孵化器方法，可催化数据驱动的研究并创建准备解决海岸复杂问题的研究人员网络。该计划以其他“科学冲刺”为基础，其中的研究人员团队组装，将一个想法转变为简短，固定的时间窗口内的开放，自由访问的研究产品，从而加快了科学进步。该项目将以三种方式推进科学：1）通过使用数据驱动的方法来解决沿海跨学科问题的科学家组成； 2）每个事件中的科学家都会创建开放的工具，代码，可交付成果和数据产品，从而创建免费的方法和知识； 3）事件之间的多次事件和迭代将使Sprint方法评估，并在海岸线和人们的交集中成功地生产科学。这三个Sprint事件专注于使用开放数据和机器学习和机器学习和机器学习和将利用data.gov和其他公平（可访问，可互操作，可重新使用）来源可用的大量数据量。目的是迅速产生结果和可交付成果。科学界的申请将针对这三个计划中的每个事件进行征集，并且选择队列将优先考虑各种领域和观点。每个事件都将遵守行为代码，该守则还将包括在Sprint上生成的代码，数据和报告的“默认情况下打开”的语句。在每个活动中，参与者都会分为小组，花72个小时从事选定的项目。小组将在每个事件结束时产生口头和书面报告以及关联的开源代码。每个事件的结果将使用调查（事前和事件后）以及遵循与每个事件中的开放交付物相关联的数字对象标识符的使用。参与者使用的数据集也将在公开可用的网站上收集并策划，作为众包相关的开放数据列表。这三个冲刺活动将在北卡罗来纳州和科罗拉多州举行。一系列外部合作者将与参与者进行互动和建立联系。该奖项反映了NSF的法定任务，并通过使用基金会的知识分子优点和更广泛的影响标准来评估值得支持。

项目成果

期刊论文数量（11）

专著数量（0）

科研奖励数量（0）

会议论文数量（0）

专利数量（0）

A Reproducible and Reusable Pipeline for Segmentation of Geoscientific Imagery

DOI：
10.1029/2022ea002332
发表时间：
2022-09-01
期刊：
EARTH AND SPACE SCIENCE
影响因子：
3.1
作者：
Buscombe, D.;Goldstein, E. B.
通讯作者：
Goldstein, E. B.

An Active Learning Pipeline to Detect Hurricane Washover in Post-Storm Aerial Images

用于检测风暴后航空图像中飓风冲刷的主动学习管道

DOI：
10.31223/x5jw23
发表时间：
2020
期刊：
AI for Earth Sciences Workshop at NeurIPS 2020
影响因子：
0
作者：
Goldstein EB, Mohanty SD
通讯作者：
Goldstein EB, Mohanty SD

Human–coastal coupled systems: Ten questions

人类与海岸耦合系统：十个问题

DOI：
10.1017/cft.2023.8
发表时间：
2023
期刊：
Cambridge Prisms: Coastal Futures
影响因子：
0
作者：
McNamara, Dylan E.;Lazarus, Eli D.;Goldstein, Evan B.
通讯作者：
Goldstein, Evan B.

psi-collect: A Python module for post-storm image collection and cataloging

psi-collect：用于风暴后图像收集和编目的 Python 模块

DOI：
10.21105/joss.02075
发表时间：
2020
期刊：
Journal of Open Source Software
影响因子：
0
作者：
Moretz, Matthew;Foster, Daniel;Weber, John;Chowdhury, Rinty;Rafique, Shah;Goldstein, Evan;Mohanty, Somya
通讯作者：
Mohanty, Somya

Prototyping a collaborative data curation service for coastal science

沿海科学协作数据管理服务原型