The Learner Data Institute: Harnessing The Data Revolution To Make The Learning Ecosystem More Effective, Efficient, and Engaging

该项目将为Learner Data Institute奠定基础，进一步了解人们是如何学习的，如何改进适应性教学系统，以及如何创建更有效、更高效、更吸引人和负担得起的学习系统。一个由学术界、产业界和政府组成的跨学科团队将共同努力，提高学习生态系统的有效性，造福数十万学生，让教师为未来的学习生态系统做好准备，并通过使用大数据和云计算，为加速发现和转变教育生态系统做出贡献。该项目将影响多个社区，包括学习科学、数据科学、教育中的人工智能、评估、教育数据挖掘和机器学习。这些成果将通过培训材料、讲习班、教程以及为研究人员和从业人员开设的课程广泛传播。Learner Data Institute为期两年的概念化阶段将专注于建立强大的研究人员社区，确定研究优先事项，并开发跨学科原型解决方案，以应对关键的学生学习、网络学习和学习工程挑战。基于现代学习理论和教育技术、人工智能、传感技术、信号处理和数据科学的最新进展，该团队将在多方面(认知、动机、情感等)探索新的前沿。学习者数据收集、分析和可视化，以了解并可能改变学习者使用技术学习的方式。多学科团队将解决核心研究问题，例如：(1)如何将分布广泛的跨学科研究人员、开发人员和实践者群体转变为实践社区，能够充分利用数据革命，造福于学习者和教育利益相关者；(2)如何将自适应教学系统(AISS)和数据科学用作研究工具，以进一步了解学习者是如何学习的；(3)如何利用人-技术与数据和数据科学的伙伴关系来提高学习者的水平？老师呢？有能力以促进学习的方式应用技术，并提高AISS的有效性、可扩展性和可负担性，以最大限度地发挥未来学习生态系统的潜力；以及(5)更一般地，如何扩展数据科学的前沿以包括：新的数据收集和设计方法；更具解释性、知识含量更丰富的机器学习方法(例如结合深度学习和马尔可夫逻辑)；可扩展的新推理和学习算法；数据中的对称性和联合相关性；以及从非结构化、半结构化和结构化数据中识别因果机制的方法。虽然该项目将在在线和混合学习环境中处理核心教育任务，但拟议的数据科学方法和模式一般适用于其他教学环境以及其他科学和工程领域。项目中使用的所有模型、软件、流程和数据都将被记录和传播，供每个人使用和建立项目成果。该项目是国家科学基金会利用数据革命大创意活动的一部分。这项工作由教育和人力资源局共同资助。这一奖项反映了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。

The project will lay the foundation for a Learner Data Institute, to further understanding of how people learn, how to improve adaptive instructional systems, and how to create learning systems that are more effective, efficient, engaging, and affordable. An interdisciplinary team of people from academia, industry, and government will work together to improve the effectiveness of the learning ecosystem for the benefit of its hundreds of thousands of students, prepare teachers for the future learning ecosystem, and contribute to accelerating discovery and transforming the education ecosystem through the use of big data and cloud computing. The project will impact a number of communities including learning sciences, data science, artificial intelligence in education, assessment, educational data mining, and machine learning. The outcomes will be widely disseminated through training materials, workshops, tutorials, and a course for researchers and practitioners. The two-year conceptualization phase of the Learner Data Institute will focus on building a strong community of researchers, define research priorities, and develop interdisciplinary prototype solutions that address critical student learning, cyber-learning, and learning engineering challenges. Based on modern theories of learning and recent advances in educational technologies, artificial intelligence, sensing technologies, signal processing, and data science, the team will explore new frontiers in multi-faceted (cognitive, motivational, emotional, etc.) learner data collection, analysis, and visualization in order to understand and possibly transform how learners learn with technology. The multi-disciplinary team will address core research questions such as: (1) how to transform a widely distributed group of interdisciplinary researchers, developers, and practitioners into a community of practice that can fully exploit the data revolution for the benefit of the learners and educational stakeholders; (2) how adaptive instructional systems (AISs) and data science can be used as a research vehicle to further understanding of how learners learn; (3) how the human-technology partnership with data and data science can be used to improve learners? and teachers? ability to employ technology in ways that facilitate learning and improve the effectiveness, scalability, and affordability of AISs in order to maximize the potential of learning ecologies of the future; and (5) more generally, how to extend the frontiers of data science to include: new methods of data collection and design; more interpretable, knowledge-rich machine learning methods (e.g., by combining Deep Learning with Markov Logic); scalable new inference and learning algorithms symmetries and joint dependencies in the data; and methods for identifying causal mechanisms from unstructured, semi-structured, and structured data. While the project will address core educational tasks in the context of online and blended learning environments, the proposed data science methods and models are generally applicable to other instructional contexts as well as other science and engineering areas. All models, software, processes, and data used in the project will be documented and disseminated for everyone to use and build on the outcomes of the project.This project is part of the National Science Foundation's Harnessing the Data Revolution Big Idea activity. The effort is jointly funded by the Directorate for Education and Human Resources.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.