SCC-PG: Early Community Intervention for Neighborhood Revitalization Using Artificial Intelligence and Emerging Technologies

美国城市社区的住房弃置给居民和市政府带来了深刻的挑战。在堪萨斯城，超过1万所房屋被废弃，每幢房屋的安全拆除成本为1万美元。然而，拆迁进一步削弱了弱势社区的物质和社会结构。该提案授权社区成员、城市领导人、技术专家和当地公立学校学生在规划过程中利用人工智能（AI）。该提案利用与市政府的合作，在城市车辆上使用摄像头来识别房屋腐烂的早期标志。其目的是建立一个早期预警系统，地方政府和社区协会可以提供微观政策投入，以便停止放弃的进程。该项目组建了一个由数据科学家、社区领袖和社会科学家组成的多学科团队，为解决废弃住房问题奠定了科学和工程基础。在这项规划拨款中，该项目集中在密苏里州堪萨斯城（KCMO）的一个小社区，开发减少住房废弃问题所需的工具和规划。具体而言，该项目将重点关注艾芬豪社区，该社区约有5,500名低收入少数民族居民（95%是非裔美国人，35%低于贫困水平），其中约40%的土地被空置土地和废弃房产占用。该项目的具体目标是：1)通过应用深度学习技术和统计建模来提高对住房遗弃的科学认识；2)建立一个多学科和多样化的研究社区，开发测量、预测和预防住房遗弃蔓延的工具；3)将KCMO的社区利益相关者整合到一系列社区项目中，以实现最佳自我控制，减少住房遗弃和解决剩余问题。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。

Housing abandonment in urban American communities presents a profound challenge for residents and city governments. In Kansas City, over 10,000 houses have been abandoned, with safe demolition costing $10,000 per house. However, demolition further weakens the physical and social fabric of disadvantaged neighborhoods. This proposal empowers community members, city leaders, technologists, and local public-school students in a planning process to leverage the use of Artificial Intelligence (AI). The proposal leverages collaboration with the municipal government to use cameras on city vehicles to identify early markers of housing decay. The intention is to develop an early warning system where local government and neighborhood associations can provide micro-policy inputs in order to halt the process of abandonment.This project has formed a well-integrated multidisciplinary team of data scientists, community leaders and social scientists to lay out the scientific and engineering foundations for addressing the issue of abandoned housing. In this planning grant, the project concentrate on a small neighborhood of Kansas City, Missouri (KCMO) to develop the tools and planning required for reducing the issue of housing abandonment. Specifically, the project will focus on the Ivanhoe neighborhood, which is home to around 5,500 mostly low-income minority residents (95% African-American, 35% below the poverty level) and which has about 40% of its land occupied by vacant lots and abandoned properties. Specific objectives of this project are i) enhancing scientific knowledge of housing abandonment by applying deep learning technology and statistical modeling, ii) fostering a multidisciplinary and diverse research community to develop tools for measuring, predicting, and preventing the spread of housing abandonment, and iii) integrating KCMO’s community stakeholders into a series of community projects for optimal self-control to reduce abandoned housing and resolve residual problems.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.