Crime prevention requires accurate prediction of the spatial and temporal distribution of criminal activities to effectively allocate law enforcement resources. However, many trending crime prediction algorithms lack comprehensive spatio-temporal structures and often consider only single input variables. This study innovatively using in ST-Cokriging method integrated both historical crime records as the primary variable and crime-related geo-tagged Twitter data as the co-variable for crime prediction. The predictive method has been specifically developed to assess crime risk across three major crime types—street crime, property crime, and vehicle crime—and applied in the San Francisco Bay Area (SFBA), California, a region characterized by high development and heightened crime sensitivity, for both prediction and validation. The results indicate that incorporating social media data into a spatio-temporal statistical method improves the associations between predicted and actual crime risk, reduced the Root Mean Squared Error (RMSE), and enhanced the identification of crime risk areas for both weekdays and weekends across three crime types compared to the method without the co-variable. This study presents a new multi-variable approach to more accurately predict crime, enabling law enforcement proactively address crime of varying nature in urban areas.
Sentiment expressions related to child maltreatment (CM) in public discourse are influenced by demographic, economic, and cultural factors and individual characteristics. Using 188,429 geotagged CM-related tweets during 2018–2022, we explored public sentiment expression about CM across the contiguous U.S. We applied multiscale geographically weighted regression (MGWR) to examine how contextual factors relate to the percentage of CM-related tweets with negative sentiment at the county level, revealing the spatial heterogeneity and varying geographic scales of these associations. Counties with higher male-to-female ratios and lower education levels tended to express negative sentiment in CM-related tweets, with consistent patterns observed nationwide. Five factors exhibited spatially varying associations by U.S. region, with higher levels of negative sentiment in the following contexts: a lower percentage of residents living in group quarters or a higher percentage of same-sex couples (Eastern and Central); fewer households lacking broadband access (Central); a higher percentage of single-parent households (New England and Southern Mississippi River); and areas where professionals are mandated to report CM (Great Lakes and Southern Appalachian Mountains). This study provides critical insights for policymakers to adjust policies, educators to design focused interventions, and the public to raise CM awareness. The methodology also provides a valuable framework for investigating public discourse on other social issues.