Proto-OKN Theme 1: DREAM-KG: Develop Dynamic, REsponsive, Adaptive, and Multifaceted Knowledge Graphs to address homelessness with Explainable AI

该项目旨在创建一个知识图谱，全面了解导致无家可归的社会、经济和政治因素，同时对现有服务和资源进行分类，以支持无家可归的人。该系统的主要用户将是无家可归者、前线案件工作者、执法人员、非营利组织和联邦机构。知识图谱将为用户提供必要的见解，以更有效地解决无家可归问题。这包括理解核心结构因素、有效的干预方法、社区资源、不同背景的文化差异，以及从地方到联邦层面的政策知识。该项目将利用新兴的拓扑数据分析工具、人工智能（AI）模型以及与无家可归相关的数据采集和分析的创新方法。由此产生的知识图谱将具有用户友好的用户界面，提供易于访问的数据和见解，加强对无家可归者的服务提供。通过利用几何理论、可解释的人工智能和本体技术，该项目旨在促进对人工智能应用的理解，弥合人工智能与无家可归研究之间的差距，并为预防无家可归提供新的解决方案。该项目的教育部分包括指导代表性不足的群体，并将项目成果纳入各种课程，如拓扑数据分析、数据语义和社会工作教育。项目成果将广泛共享，包括受影响社区、临床医生、政策制定者和研究人员在内的多个利益相关者将参与其中，旨在解决全国无家可归危机，改善资源分配和住房政策。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。

This project aims to create a knowledge graph that will provide a comprehensive understanding of the social, economic, and political factors that contribute to homelessness while triaging existing services and resources to support people experiencing homelessness. The primary users of the system will be people experiencing homelessness, front-line case workers, law enforcement agents, non-profit organizations, and federal agencies. The knowledge graph will equip users with essential insights to address the homelessness problem more effectively. This includes understanding core structural factors, effective intervention methods, community resources, cultural nuances across diverse backgrounds, and policy knowledge from local to federal levels. The project will utilize emerging topological data analysis tools, artificial intelligence (AI) models, and innovative approaches for data acquisition and analysis related to homelessness. The resulting knowledge graph will feature user-friendly user interfaces, offering easy access to data and insights, enhancing service delivery to homeless populations. By harnessing geometric theory, explainable AI, and ontology techniques, the project aims to advance understanding in AI applications, bridge the gap between AI and homelessness research, and offer novel solutions for homelessness prevention. The educational component of the project includes mentoring underrepresented groups and incorporating project outcomes into various courses, such as Topological Data Analysis, Data Semantics, and Social Work education. The project outcomes will be shared broadly, engaging multiple stakeholders including the affected communities, clinicians, policymakers, and researchers, aiming to address the national homelessness crisis and improve resource allocation and housing policies.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.