Scientific Discovery from Chemical Data Analyses

化学数据分析的科学发现

• 批准号: 2011061
• 负责人: Karl Booksh
• 金额: $ 5万
• 依托单位: University of Delaware
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-09-15 至 2022-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2011061&HistoricalAwards=false
• 关键词: Scientific; Discovery; Chemical; Data; Analyses

With support from the Division of Chemistry and partial co-funding from the Divisions of Mathematical Sciences and the Office of Advanced Cyberinfrastructure, Professors Karl Booksh and Sharon Neal at the University of Delaware are organizing a workshop to identify new frontiers in data analyses associated with chemical measurements for scientific discovery. Advances in scientific instrumentation have helped enable the "data revolution" through collection of vast amounts of data, both to probe a single sample in exquisite detail and to collect snippets of data from a wide array of interesting samples. Data analytic strategies, often under the banner of machine learning and artificial intelligence for example, offer new tools to help researchers gain valuable scientific insights from such data collections. Unfortunately, there remains an adaptation gap between these tools and researchers with application domain knowledge who would benefit from the availability of more advanced data analysis methods. This workshop seeks to identify convergent areas of chemical research that would most benefit from new cutting-edge tools for data analysis and to identify means of reducing the impediments to broad and rapid adoption of such tools by researchers with relevant application domain knowledge. The workshop will also explore needs and means for training of students in valuable data analysis skills. The workshop will bring together researchers in the chemical sciences and related areas (e.g., environmental analysis, biological chemistry, materials science, and national security monitoring) and experts in novel data analytics (e.g., machine learning, artificial intelligence, graph theory). Participants will explore the relationship between pressing scientific questions in convergent application fields and the potential for new analytical methods that uniquely address these questions. For example, how might graph theory or network analyses advance prediction of new catalysts and more efficient reaction sequences? What insights might machine learning or artificial intelligence derive from high-resolution, multivariate images of catalytic surfaces? How can sensor arrays be best adapted to improve understanding and prediction of the chemistry of dynamic environmental processes? Participants will strive to identify where collaborations are necessary to advance convergence research and means of facilitating establishment of those collaborations. The workshop will also consider the education and training needs for creating a research workforce versed in both measurement science and data analysis. The workshop will engage approximately 30 thought-leaders spanning academic, industrial, and governmental sectors using virtual and/or in-person formats.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.

在化学部的支持下，并在数学科学部和高级网络基础设施办公室的部分共同资助下，特拉华州大学的卡尔·布克什教授和莎伦·尼尔教授正在组织一个研讨会，以确定与科学发现的化学测量相关的数据分析的新前沿。 科学仪器的进步有助于通过收集大量数据来实现“数据革命”，既可以对单个样本进行细致的探测，也可以从大量有趣的样本中收集数据片段。 数据分析策略，通常是在机器学习和人工智能的旗帜下，例如，提供新的工具来帮助研究人员从这些数据收集中获得有价值的科学见解。 不幸的是，这些工具和具有应用领域知识的研究人员之间仍然存在适应差距，这些研究人员将从更先进的数据分析方法中受益。 该研讨会旨在确定最能受益于新的尖端数据分析工具的化学研究的趋同领域，并确定减少具有相关应用领域知识的研究人员广泛和快速采用此类工具的障碍的方法。 讲习班还将探讨培训学生掌握宝贵的数据分析技能的需要和手段。讲习班将汇集化学科学和相关领域的研究人员（例如，环境分析、生物化学、材料科学和国家安全监测）和新数据分析方面的专家（例如，机器学习、人工智能、图论）。 与会者将探讨聚合应用领域中紧迫的科学问题与独特解决这些问题的新分析方法的潜力之间的关系。 例如，图论或网络分析如何推进对新催化剂和更有效的反应序列的预测？ 机器学习或人工智能可以从催化表面的高分辨率、多变量图像中获得什么样的见解？ 传感器阵列如何最好地适应，以提高动态环境过程的化学的理解和预测？ 与会者将努力确定在哪些方面需要合作，以推进趋同研究，并确定促进建立这些合作的手段。 讲习班还将审议教育和培训需求，以建立一支精通测量科学和数据分析的研究队伍。 该研讨会将采用虚拟和/或面对面的形式邀请学术界、工业界和政府部门的约30位思想领袖参加。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。