EXP: Collaborative Research: Extracting Salient Scenarios from Interaction Logs (ESSIL)

EXP：协作研究：从交互日志中提取显着场景 (ESSIL)

• 批准号: 1623091
• 负责人: Andee Rubin
• 金额: $ 32.5万
• 依托单位: TERC Inc
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2016
• 资助国家: 美国
• 起止时间: 2016-09-01 至 2020-12-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1623091&HistoricalAwards=false
• 关键词: EXP; Collaborative; Research; Extracting; Salient

The Extracting Salient Scenarios from Interaction Logs (ESSIL) project proposes to develop a new type of educational technology to support students' learning about complex systems from their participation in a multi-person immersive simulation. Many important challenges we face today as a society -- including responding to climate change, managing global economies, city planning, disease outbreaks -- are "complex systems" problems, meaning that important phenomena in each (for instance trends in weather, stock bubbles, traffic jams, disease transmission) result not from a single cause, but because many small causes combine together. Participating in a simulation has the potential to help students understand the principles of complex systems, but because different principles surface depending on how each simulation unfolds, it can be difficult for teachers to adjust their lesson plans on the fly to highlight the principles that emerge in a given simulation run. To address this challenge, ESSIL will develop methods to create "automatic salient recaps," as a way to help learners and their teachers make better sense of simulations. These recaps, which will be automatically generated, provide a story of "what happened" in the simulation in a way that both helps students remember their experience and reveals important scientific principles. Teachers and other facilitators will use these recaps, along with an accompanying discussion guide, to support productive learning conversations about the scientific principles incorporated in a simulation. The recaps will be developed for a large-scale immersive simulation installed at the New York Hall of Science (NYSCI), potentially improving the educational experience of thousands of daily visitors. The capabilities developed to produce them have widespread applicability, because logs of student interactions are routinely produced by many educational systems. The project is supported by the Cyberlearning and Future Learning Technologies Program, which funds efforts that will help envision the next generation of learning technologies and advance what we know about how people learn in technology-rich environments. Cyberlearning Exploration (EXP) Projects explore the viability of new kinds of learning technologies by designing and building new kinds of learning technologies and studying their possibilities for fostering learning and challenges to using them effectively. The immersive simulation context for the project is Connected Worlds, an embodied, multi-person ecology simulation at NYSCI, with pedagogical goals around sustainability and systems thinking. Using logs from groups of students interacting with Connected Worlds, ESSIL will construct selective recaps of their experience that both are personally salient to them (by including memorable details of their experience) and have explanatory coherence (to enable their discussion of important interconnections in the simulation's underlying model). Artificial Intelligence-based methods will be developed to 1) identify salient changes in the state of the simulation during student interaction and 2) construct qualitative models of causal chains that could have led to these changes. These qualitative models will be used to generate salient recaps and discussion guides based on them, which will be provided to teachers whose classes are visiting NYSCI. The effectiveness of the innovation will be investigated by comparing visiting students' conversations with and without ESSIL-generated discussion supports and by interrogating their resulting models of the Connected Worlds system through concept maps.

从交互日志中提取突出场景（ESSIL）项目提出开发一种新型教育技术，以支持学生通过参与多人沉浸式模拟来学习复杂系统。作为一个社会，我们今天面临的许多重要挑战——包括应对气候变化、管理全球经济、城市规划、疾病暴发——都是“复杂系统”问题，这意味着每个系统中的重要现象（例如天气趋势、股票泡沫、交通拥堵、疾病传播）不是由单一原因造成的，而是许多小原因综合在一起造成的。参与模拟有可能帮助学生理解复杂系统的原理，但由于不同的原理取决于每个模拟的展开方式，教师很难在飞行中调整他们的课程计划，以突出在给定的模拟运行中出现的原理。为了应对这一挑战，ESSIL将开发创建“自动突出重述”的方法，以帮助学习者和他们的老师更好地理解模拟。这些概述将自动生成，提供模拟中“发生了什么”的故事，以一种既帮助学生记住他们的经历又揭示重要科学原理的方式。教师和其他促进者将使用这些概述，以及随附的讨论指南，以支持有关模拟中包含的科学原理的生产性学习对话。这些概述将用于安装在纽约科学馆（NYSCI）的大规模沉浸式模拟，有可能改善每天成千上万游客的教育体验。为生成这些日志而开发的功能具有广泛的适用性，因为许多教育系统通常都会生成学生交互日志。该项目由网络学习和未来学习技术项目支持，该项目资助的工作将有助于设想下一代学习技术，并进一步了解人们在技术丰富的环境中如何学习。网络学习探索（EXP）项目通过设计和构建新型学习技术并研究其促进学习的可能性和有效使用它们的挑战来探索新型学习技术的可行性。该项目的沉浸式模拟环境是连接世界，这是NYSCI的一个具体化的多人生态模拟，其教学目标是围绕可持续性和系统思维。使用来自学生群体与互联世界互动的日志，ESSIL将对他们的经历进行选择性回顾，这些经历对他们个人来说都是突出的（通过包括他们经历的难忘细节），并且具有解释性的一致性（使他们能够讨论模拟底层模型中的重要互连）。将开发基于人工智能的方法，以1)识别学生互动过程中模拟状态的显著变化，2)构建可能导致这些变化的因果链的定性模型。这些定性模型将用于生成突出的概述和讨论指南，这些概述和讨论指南将提供给访问NYSCI的班级的教师。通过比较访问学生在使用和不使用essil生成的讨论支持的情况下的对话，以及通过概念图询问他们对互联世界系统的最终模型，将对创新的有效性进行调查。