Information fusion to estimate resilience of dense urban neighborhoods

Abstract

Diverse sociocultural influences in rapidly growing dense urban areas may induce strain on civil services and reduce the resilience of those areas to exogenous and endogenous shocks. We present a novel approach with foundations in computer and social sciences, to estimate the resilience of dense urban areas at finer spatiotemporal scales compared to the state-ofthe-art. We fuse multi-modal data sources to estimate resilience indicators from social science theory and leverage a structured ontology for factor combinations to enhance explainability. Estimates of destabilizing areas can improve the decision-making capabilities of civil governments by identifying critical areas needing increased social services.Keywords: Dense urban areas, behavior modeling, social/cultural modeling, social capital, situation/threat assessment INTRODUCTIONTwo-thirds of the world's population is expected to live in urban areas by the year 2050. This rapid growth of dense neighborhoods with diverse sociocultural influences may strain civil services, e.g., law enforcement, health care, and waste management, reducing the resilience of these areas to exogenous and endogenous shocks. Aggregate statistics for large, sprawling, urban areas are likely to miss critical indicators.To address these issues, we developed a novel approach that advances the state-of-the-art by (i) ascertaining the stability and resilience of densely populated areas at much finer geospatial and temporal scales, and (ii) basing that estimate on a solid social science foundation. We fuse multi-modal data sources to estimate the prevalence of indicators from social theories (e.g., social capital). Aggregate statistics from structured city-level data (censuses, surveys) provide a baseline for the entire city, while social media and news reports provide features at shorter time scales, and smaller geographic areas (i.e., neighborhoods, city blocks) introducing deviations from the baseline. This social science foundation, combined with our structured ontology, leads to enhanced explainability of the stability assessment, increasing confidence in downstream decisions. Knowledge of destabilizing areas informs the decisionmaking process in civil governments by identifying neighborhoods in need of social services and civil control. Our computational framework estimates the capacity of a dense urban area to cope with stress and volatility, allowing governments, soldiers, humanitarian organizations, and private companies to better understand and manage local disturbances. THEORETICAL FRAME: SOCIAL SCIENCE FOUNDATIONThere is a need to bridge the gap between the computational and social sciences. Modern decision-makers require support from computational tools with near real-time feedback, however social science theories are still validated/quantified after lengthy surveys and structured interviews. We performed comprehensive research and built methods using open-source data and text analytics to allow data science to quantify social science theories at a rapid pace. As a first step...

Correspondence between spectral reflectance and features of the built environment for community resilience

“…The findings in this work also expand the research on infrastructure-based indicators of resilience 16 . Although there are many measures for disaster resilience, 31 our approach focuses on social capital derived from infrastructure.…”

“…The findings in this work also expand the research on infrastructure-based indicators of resilience. 16 Although there are many measures for disaster resilience, 31 our approach focuses on social capital derived from infrastructure. Previous studies have suggested that certain features embedded within the built environment serve as important mechanisms for fostering social relationships.…”

“…Most studies rely on intensive in situ research methods that, by necessity, focus on one or two small neighborhoods; however, by combining computational tools and social science concepts, researchers have been able to implement a novel approach to assess the resiliency of an entire DUT. [16][17][18] Palladino et al 16 used an urban resilience ontology that related social structures and social events including access and capacity to compute a social capital measure for the entirety of a DUT landscape. Their work leveraged a suite of open-source spatial infrastructure, land-cover, and development data from sources such as OpenStreetMap (OSM) 19 and Google Places; 20 however, there is no guarantee of completeness for these and other types of open data sources, and the accuracy of their social capital measurements relied heavily on the integrity of their input datasets.…”

“…Naturally, this resource varies depending on a suite of demographic and infrastructural factors. Most studies rely on intensive in situ research methods that, by necessity, focus on one or two small neighborhoods; however, by combining computational tools and social science concepts, researchers have been able to implement a novel approach to assess the resiliency of an entire DUT 16 – 18 Palladino et al 16 .…”

“…Most studies rely on intensive in situ research methods that, by necessity, focus on one or two small neighborhoods; however, by combining computational tools and social science concepts, researchers have been able to implement a novel approach to assess the resiliency of an entire DUT 16 – 18 Palladino et al 16 . used an urban resilience ontology that related social structures and social events including access and capacity to compute a social capital measure for the entirety of a DUT landscape.…”