SCC-PG:MAPPING INSTABILITY: Building an Intelligent Community Agent Platform for Understanding the Impact of Large Scale Crisis on Small Town Communities

2019冠状病毒病大流行继续以可察觉和不可察觉的方式影响建筑环境和社区生活。大城市的居住、工作和流动依赖于大型基础设施，而小社区的生活则严重依赖于个人资源和自我维持的结构。应对重大危机的一个重要步骤是及时收集丰富的数据，以了解社区的问题、斗争和需求。然而，传统的数据收集方法，如公开会议，是无效的，而且那些可能有儿童保育或工作冲突的人很少参加。在线数据收集方法，如调查，通过消除实际存在的需要，扩大了外延和包容性，但它们并不总是支持对话交流，鼓励人们对他们的问题和需求提供更深入的见解。该项目将通过设计、构建和评估对话代理来收集有关大流行对马萨诸塞州西部阿默斯特、霍利奥克和皮茨菲尔德居民影响的数据，从而改善公民数据收集。虽然这些社区在地理上很接近，但它们的人口结构、经济繁荣程度和获得公共服务的机会各不相同。这项研究有助于识别弱势、服务不足和代表性不足的群体，以分配和优先分配资源和材料，并作为解决美国小城镇类似问题的概念证明。该项目使地方官员能够充分了解这些小城镇在面对当前流行病时所面临的挑战。通过使用以人为本的方法，该项目将建立一个基于人工智能的对话代理，以收集有关社区生活的各个方面的数据，如居住、交通、工作、教育和医疗保健。这项研究将从五个方面改变计算机科学和体系结构领域的行动和运作模式：(1)通过设计和构建以社区为中心的对话代理平台来改善公共数据收集的现状；(2)实证研究不同人口统计数据与对话代理的交互方式；(3)有助于认识对话代理在收集公民有意义的输入方面的作用。(4)提供对当前流行病对小镇居民生活影响的丰富理解，这将使建筑领域了解社会、经济和文化力量对建筑环境的影响；(5)使建筑分析更加同步，并对影响它的力量作出反应。该奖项反映了美国国家科学基金会的法定使命，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。

The COVID-19 pandemic continues to impact the built environment and communities’ lives in perceptible and imperceptible manners. Unlike big cities, where dwelling, working, and mobility rely on large infrastructures, life in smaller communities relies heavily on individual resources and self-sustained structures. A vital step in responding to major crises is the timely collection of rich data to understand the community’s issues, struggles, and needs. However, traditional data collection methods such as public meetings are ineffective and poorly attended by those who may have childcare or work conflicts. Online data collection methods such as surveys broaden the outreach and inclusivity by eliminating the need for physical presence, but they do not always support a conversational exchange that encourages people to provide deeper insights into their issues and needs. This project will improve civic data collection by designing, building, and evaluating a conversational agent to collect data about the pandemic’s impact on residents of Amherst, Holyoke, and Pittsfield in Western Massachusett. While these communities are in close geographic proximity, they have different demographics, economic prosperity, and access to public services. This research facilitates identifying vulnerable, under-served, and under-represented groups for allocation and prioritization of resources and materials and serves as a proof of concept for addressing similar issues for small towns across the United States.This project enables local officials to gain a rich understanding of those small towns’ challenges in the face of the current pandemic. By using human-centered methods, this project will build an AI-based conversational agent to collect data about diverse aspects of communities' lives such as Dwelling, Transportation, Work, Education, and Healthcare. This research will transform modes of action and operation in both Computer Science and Architecture fields in five ways: (1) Advancing the status quo of public data collection by designing and building a community-centered conversational agent platform, (2) Empirically addressing how various demographics interact with conversational agents, (3) Contributing to recognition of conversational agents’ role in gathering meaningful input from citizens, (4) Providing a rich understanding of the impacts of the current pandemic on small-town residents’ lives that will inform the architecture field about the effects of social, economic, and cultural forces on the built environment and (5) Making architectural analysis more synchronous and responsive to forces that affect it.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.