Data-Driven, Human-in-the-Loop Support for Facilitating Participatory Learning Activities

博物馆越来越多地开发数字学习体验，参观者可以生成有趣的使用数据，但这些数据很少用于支持学习。该项目提出了一种“人在环中”的促进模式，博物馆专家可以通过挖掘展品使用数据获得数据可视化和其他反馈。这些视频将被用作辅导员的指导支持，因为它们帮助参观者理解数字展品现在提供的丰富学习体验。虽然这种方法与博物馆学习直接相关，但人在环中系统代表了对网络学习系统的新看法，该系统将适用于许多其他正式和非正式的STEM学习环境，这些学习环境需要在专家观察者的指导下进行即时、及时的反思。干预将用于一个开放式的参与式展览，该展览一次邀请多达50名参观者操纵模拟生态系统，目的是让参观者通过探索新出现的现象来参与与可持续发展和复杂系统相关的概念。开放式学习体验在博物馆展品中很常见，但在教育数据挖掘(EDM)社区中还没有得到充分的研究。这项研究将增加关于开放式问题的日益增长的EDM文献，并提供关于如何设计博物馆展品以提供可用于分析和数据挖掘的数据的见解。这项研究的一个目标是开发和提炼一种信息形式的类型学，以便博物馆和其他使用实时数据的促进者能够理解并采取行动，可以用来为未来促进学习的工具的分析和软件设计提供信息。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。

Museums are increasingly developing digital learning experiences where visitors generate interesting usage data, but these data are rarely used to support learning. This project poses a "human-in-the-loop" model of facilitation where museum experts are provided with data visualizations and other feedback derived by mining exhibit use data. These visuals will be used as coaching support for facilitators as they help visitors make sense of the rich learning experiences that digital exhibits now offer. While this approach is immediately relevant to museum learning, human-in-the-loop systems represent a new take on cyberlearning systems that will be applicable to many other formal and informal STEM learning environments that demand immediate, just-in-time reflection that is guided by an expert observer.The intervention will be used in the context of an open-ended, participatory exhibit that invites up to 50 visitors at a time to manipulate a simulated ecosystem, with the goal of having visitors engage with concepts related to sustainability and complex systems by exploring emergent phenomena. Open-ended learning experiences are common to museum exhibits but are under-studied in the Educational Data Mining (EDM) community. This research will add to the growing EDM literature on open-ended problems and provide insights for how museum exhibits can be designed to supply data that can be useful for analytics and data mining. A goal of this research is to develop and refine a typology of the forms of information that museum and other facilitators working with real-time data can make sense of and act on, which could be used to inform both the analytics and the software design for future tools to facilitate learning.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.