A fully Bayesian tracking algorithm for mitigating disparate prediction misclassification
In: International journal of forecasting, Band 39, Heft 3, S. 1238-1252
ISSN: 0169-2070
2 Ergebnisse
Sortierung:
In: International journal of forecasting, Band 39, Heft 3, S. 1238-1252
ISSN: 0169-2070
In: Policing: a journal of policy and practice, Band 17
ISSN: 1752-4520
AbstractGunshot detection technology (GDT) is expected to impact gun violence by accelerating the discovery and response to gunfire. GDT should further collect more accurate spatial data, as gunfire is assigned to coordinates measured by acoustic sensors rather than addresses reported via 9-1-1 calls for service (CFS). The current study explores the level to which GDT achieves these benefits over its first 5 years of operation in Kansas City, Missouri. Data systems are triangulated to determine the time and location gunfire was reported by GDT and CFS. The temporal and spatial distances between GDT and CFS are then calculated. Findings indicate GDT generates time savings and increases spatial precision as compared to CFS. This may facilitate police responses to gunfire events and provide more spatially accurate data to inform policing strategies. Results of generalized linear and multinomial logistic regression models indicate that GDT benefits are influenced by a number of situational factors.