A5079717522- Air Quality and Health Impacts
- Topic Modeling
- Air Quality Monitoring and Forecasting
- Natural Language Processing Techniques
- Atmospheric chemistry and aerosols
- Climate Change and Health Impacts
- Impact of Light on Environment and Health
- Human Mobility and Location-Based Analysis
- Semantic Web and Ontologies
- Mobile Crowdsensing and Crowdsourcing
- Innovative Human-Technology Interaction
- Data Quality and Management
- Text and Document Classification Technologies
- Environmental Justice and Health Disparities
- Ethics and Social Impacts of AI
- Online Learning and Analytics
Microsoft (United States)
2022-2025
Transforming unstructured text into structured and meaningful forms, organized by useful category labels, is a fundamental step in mining for downstream analysis application. However, most existing methods producing label taxonomies building text-based classifiers still rely heavily on domain expertise manual curation, making the process expensive time-consuming. This particularly challenging when space under-specified large-scale data annotations are unavailable. In this paper, we address...
This paper presents Eclipse, a platform for low-cost urban environmental sensing using solar-powered and cellular-connected devices. Dense sensor networks promise to monitor pollution at fine spatial temporal resolutions, yet few cities have actually implemented such due high costs limited accuracy. We address these barriers by developing an end-to-end framework air quality with minimal infrastructure requirements. designed unobtrusive device that collects data on particulate matter (PM <inf...
Understanding user intents in information access scenarios can help us provide more relevant and personalized search results recommendations. However, analyzing is not easy, especially for emerging forms of Web such as Artificial Intelligence (AI)-driven chat. To understand from retrospective log data, we need a way to label them with meaningful categories that capture their diversity dynamics. Existing methods rely on manual or Machine-Learned (ML) labeling, which either expensive...
Objectives. To evaluate the efficacy of a novel, real-time sensor network for routine monitoring racial and economic disparities in fine particulate matter (PM2.5; ≤ 2.5 µm diameter) exposures at neighborhood level. Methods. We deployed dense low-cost PM2.5 sensors Chicago, Illinois, to associations between neighborhood-level composition variables (percentage Black residents, percentage Hispanic/Latinx households below poverty) interpolated PM2.5. Relationships were assessed spatial lag...
Abstract High-resolution air quality data products have the potential to help quantify inequitable environmental exposures over space and across time by enabling identification of hotspots, or areas that consistently experience elevated pollution levels relative their surroundings. However, when different high-resolution identify spatial sparsity ‘gold-standard’ regulatory observations leaves researchers, regulators, concerned citizens without a means differentiate signal from noise. This...
The digital divide describes disparities in access to and usage of tooling between social economic groups. Emerging generative artificial intelligence tools, which strongly affect productivity, could magnify the impact these divides. However, affordability, multi-modality, multilingual capabilities tools also make them more accessible diverse users comparison with previous forms tooling. In this study, we characterize spatial differences U.S. residents' knowledge a new AI tool, ChatGPT,...
Urban environmental monitoring campaigns depend on expertise from city agencies, residents, and researchers. Deployment efforts rarely include all three stakeholders, typically leading to initiatives that struggle produce credible, actionable data. We describe the implementation of a large-scale, long-term air quality sensing network in Chicago Illinois; detail stakeholder interviews meetings; present interfaces—–a website accessible via in-situ QR codes, APIs, mobile, mixed-media...