Leveraging Machine Learning to Explore the Effects of the Design2Data Course-based Undergraduate Research Experience

利用机器学习探索基于 Design2Data 课程的本科生研究经验的效果

• 批准号: 2315767
• 负责人: Justin Siegel
• 金额: $ 39.75万
• 依托单位: University of California-Davis
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-07-15 至 2026-06-30
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2315767&HistoricalAwards=false
• 关键词: Leveraging; Machine; Learning; Explore; Effects

This project aims to serve the national interest by engaging students in the Design to Data (D2D) program, a nationally networked biochemistry Course-based Undergraduate Research Experience (CURE). The project will help to determine which parts of the student experience positively impact graduation and access to STEM careers. D2D engages students in exploring protein science through a cutting-edge computationally driven research project. This project anchors the curriculum across a network of highly motivated, early-adopter faculty members teaching students in varied STEM disciplines and levels. The hands-on D2D learning experience prepares students for success in this new era of biology while crowdsourcing data collection for improved protein modeling artificial intelligence methods. The large, diverse D2D student body offers a rich opportunity to explore novel, machine learning-based assessment methods for CURE education research. In reaching the aims of this grant, this project will (a) pioneer a cutting-edge approach to investigating student learning in research experiences and will make these data analysis methods broadly accessible to STEM education researchers, and (b) create equitable, meaningful research experiences for thousands of students, many of whom would not have otherwise had the opportunity. D2D’s undergraduate research project anchors the program, and D2D faculty network members facilitate its implementation by integrating the project into their classes on a wide variety of campuses across the United States. Participating students functionally characterize novel enzyme mutants generating data for protein modeling stakeholders to explore with the goal of developing better functionally predictive tools for more rapid solutions to human-centered problems. Not only is there potential to meaningfully advance science through the program, but the experience enables equitable access to cutting-edge biotechnology training that is in high demand by employers. Reaching many students with this program is tractable: D2D readily integrates into lab practicum settings across the disciplines and from first-year to senior-level classes. The network includes forty institutions and will engage approximately 4,000 students over the funding period. The project objective is to use this large, diverse population in assessing mediators to psycho-social and behavioral outcomes linked to STEM persistence by collecting multi-level motivational data at scale with layered variables and benchmark cutting edge machine learning methods for the data analysis. The project activities will (a) support dedicated network continuity coordination to maintain current levels of faculty participation, and (b) assessment activities to plan and execute comprehensive data collection and machine learning (ML)-based analysis that captures and evaluates a deep set of discrete CURE-implementation variables. From these activities, meaningful professional development beyond the D2D Network will be promoted by making the cutting-edge student learning data analysis methods accessible to other education researchers. Finally, this project will put research into hands of thousands of students and enable more equitable access to CUREs, increasing the diversity of students participating in research. The NSF IUSE: EDU Program supports research and development projects to improve the effectiveness of STEM education for all students. Through its Engaged Student Learning track, the program supports the creation, exploration, and implementation of promising practices and tools.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.

该项目旨在通过让学生参与设计到数据(D2D)计划来服务于国家利益，D2D计划是一个基于全国联网的生物化学课程的本科生研究体验(CURE)。该项目将有助于确定学生经历的哪些部分对毕业和进入STEM职业生涯产生积极影响。D2D让学生通过一个尖端的计算驱动的研究项目来探索蛋白质科学。该项目通过一个高度积极的、早期采用的教职员工网络来支撑课程，在不同的STEM学科和水平上教授学生。实践D2D学习体验为学生在这个生物学的新时代取得成功做好准备，同时众包改进的蛋白质建模人工智能方法的数据收集。庞大多样的D2D学生群体为治疗教育研究提供了丰富的机会来探索新颖的、基于机器学习的评估方法。为了实现这笔赠款的目标，该项目将(A)开创一种调查学生在研究经历中的学习的前沿方法，并使STEM教育研究人员能够广泛使用这些数据分析方法，以及(B)为成千上万的学生创造公平、有意义的研究体验，否则他们中的许多人就没有机会。D2D的本科生研究项目是该计划的支柱，D2D教职员工网络成员通过将该项目整合到美国各地不同校园的课堂上来促进该计划的实施。参与的学生从功能上刻画新的酶突变的特征，为蛋白质建模利益相关者探索生成数据，目标是开发更好的功能预测工具，以更快地解决以人为中心的问题。不仅有可能通过该计划有意义地推动科学发展，而且这种经验还使雇主能够公平地获得雇主高度需求的尖端生物技术培训。通过该计划接触到许多学生是容易的：D2D很容易整合到各个学科的实验室实践环境中，以及从一年级到高级的课程。该网络包括40个机构，在资助期内将招收约4,000名学生。该项目的目标是通过收集具有分层变量的多层次激励数据和用于数据分析的基准尖端机器学习方法，在评估与STEM持续性相关的心理-社会和行为结果的中介时使用这一庞大、多样化的人群。项目活动将(A)支持专门的网络连续性协调，以维持当前的教师参与水平，以及(B)评估活动，以规划和执行全面的数据收集和基于机器学习(ML)的分析，以捕获和评估一组深入的离散治愈实施变量。通过这些活动，通过使其他教育研究人员能够访问尖端的学生学习数据分析方法，将促进D2D网络以外的有意义的专业发展。最后，该项目将把研究交到数千名学生手中，并使他们能够更公平地获得治疗，增加参与研究的学生的多样性。NSF IUSE：EDU计划支持研究和开发项目，以提高所有学生的STEM教育的有效性。通过其参与的学生学习跟踪，该计划支持有前途的实践和工具的创建、探索和实施。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。