Spatio-Temporal Crime Predictions by Leveraging Artificial Intelligence for Citizens Security in Smart Cities

Abstract

Smart city infrastructure has a significant impact on improving the quality of humans life. However, a substantial increase in the urban population from the last few years is posing challenges related to resource management, safety, and security. In order to ensure safe mobility and security in the smart city environment, this paper proposes a novel Artificial Intelligence (AI) based approach empowering the authorities to better visualize the threats and to help them identify the highly-reported crime zones yielding greater predictability of crime hot-spots in a smart city. To this end, it first investigates the Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN) to detect the hot-spots that have a higher risk of crimes to be committed. Second, for crime prediction, Seasonal Auto-Regressive Integrated Moving Average (SARIMA) exploited in each dense crime region to predict the number of crimes in the future with spatial and temporal information. The proposed HDBSCAN and SARIMA based crime prediction model is evaluated on ten years of crime data (2008-2017) for New York City (NYC). The accuracy of the model is measured by considering different time period scenarios i.e. (a) year-wise, i.e., for each year and (b) for the whole period of ten years, using an 80:20 ratio where 80\% data was used for training and 20\% data was used for testing. The proposed approach outperforms with an average Mean Absolute Error (MAE) of 11.47.

Participación comunitaria en la reducción de la inseguridad, revisión sistemática de la literatura 2016-2022

“…En la revisión sistemática de la literatura respecto a la participación comunitaria en la reducción de la inseguridad, concluida en 33 artículos, los autores coligen en que la policía no puede combatir el crimen sola, sino que necesita del apoyo de la comunidad en general (Singh and Zondi, 2020), por lo que se asumen estrategias como la prevención del delito mediante el diseño ambiental (González, 2019) o el uso de tecnologías tipo smart cities y aplicativos basados en internet (Schreurs et al, 2020) que buscan limitar el avance de la ciberdelincuencia (Butt et al, 2021). estos se suman los mecanismos de acercamiento al ciudadano, ya que el establecer un entorno seguro para las personas fomenta su desarrollo individual (Rupp et al, 2020) y lo aleja de la desigualdad (Wiedlitzka, 2021), enfatizándose la necesidad de reforzar las relaciones sociales (Singh and Zondi, 2020), en las que las fuerzas del orden deben ocupar un lugar y rol específico e importante (González, 2019); (Rukus et al, 2018); (Heinze et al, 2018); (Nussio and Norza Céspedes, 2018) en un eficiente binomio policía ciudadano (Amante et al, 2021); (Makondo et al, 2021); (Akkuş and Arslan, 2021) que reduzca la inseguridad y por ende el temor al delito, mejorando la opinión respecto a las fuerzas de policía (Akkuş and Arslan, 2021) y el nivel de denuncias por las victimas (Mbewu et al, 2021); mecanismo que recibe distintas denominaciones: (Wells and Millings, 2019); (Ceccato, 2019) como "Municipios Seguros" y "Barrios Seguros" en Chile (Hathazy, 2018), "Unidades de Policía Pacificadora" y "Pacto pela Vida" en Brasil (Wolff, 2017), "Foros de Vigilancia Comunitaria" (Makondo et al, 2021) en Sudáfrica, "Contratos de Seguridad Local" en Portugal (Amante et al, 2021) y el "Community Speedwatch" en el Reino Unido (Wells and Millings, 2019).…”

OFFICERS: Operational Framework for Intelligent Crime-and-Emergency Response Scheduling

“…Incident prediction is a vital prerequisite to effective deployment planning, as the knowledge of where emergencies occur will inform the positioning of response agents. Examples of prediction approaches used in the past include statistical aggregation of historical demands (Malleson and Andresen 2015), spatial hotspot identification with temporal regression (Butt et al 2021), risk terrain modeling (Caplan, Kennedy, and Miller 2011), continuous time modelling of burglary (Mukhopadhyay et al 2016), Gaussian Process incident generation for discrete space-time intervals (Chase et al 2019), deep learning (Chase et al 2021), (Wang et al 2017), and network analytics (Dash, Safro, and Srinivasamurthy 2018). Where many of these methods fall short is in capturing the spatio-temporal dependencies in crime data, where incidents appearing in one location or time period implies they will not also appear in the immediately adjacent locations or periods.…”