Modelo de regresión para el pronóstico de la cantidad de denuncias por delitos que se registran en las Comisarías de la Policía Nacional de Perú en Lima Metropolitana
Date
2023-08-31
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Pontificia Universidad Católica del Perú
Abstract
En nuestro país, la cantidad promedio de denuncias por delitos de manera mensual
en el año 2019 presenta 30,000 casos, lo cual se ha ido incrementando a lo largo
de los años, por lo que es de crucial importancia generar estrategias de seguridad
ciudadana que ayuden a mejorar el bienestar de la sociedad peruana, para ello es
crucial tener información fiable y de calidad para la toma de decisiones; asimismo,
tiene una Dirección de Tecnologías de la Información y Comunicación de la Policía
(DIRTIC), la cual dentro de su estructura tiene a su cargo a la División de Estadística
(DIVEST) y la División de Informática (DIVINFOR), las cuales tienen como función el
control del Sistema de Denuncias Policiales en la cual el 80 % de las comisarías de la
Policía Nacional del Perú se encuentran interconectadas y existe un 20 % que no se
encuentran interconectadas, por lo cual el 20 % de comisarías no registra información
en el Sistema de Denuncias Policiales, generando una incertidumbre para la toma
de decisiones.
Es por tal motivo que la División de Estadística (DIVEST) propone un modelo para
el pronóstico de la cantidad de denuncias por delitos que se registran en las comisarías
de la Policía Nacional de Perú en Lima Metropolitana, en base a la técnica
de regresión RANDOM FOREST REGRESSOR Y ÁRBOL DE DECISIÓN DE REGRESIÓN,
con la cual se podrá conocer cuál de las dos técnicas genera un mejor
pronóstico en base a la información que proviene del SIDPOL y del INEI.
In our country, the average number of complaints for crimes on a monthly basis in 2019 presents 30,000 cases, which has been increasing over the years, so it is of crucial importance to generate citizen security strategies that help to improve the well-being of Peruvian society, for this it is crucial to have reliable and quality information for decision-making; likewise, it has a Police Information and Communication Technologies Directorate (DIRTIC), which within its structure has in charge of the Statistics Division (DIVEST) and the Information Technology Division (DIVINFOR), whose function is to control the Police Complaints System in which 80 % of the Peruvian National Police stations are interconnected and there are 20 % that are not interconnected, for which reason 20 % of police stations do not record information in the Police Complaints System, generating uncertainty for decision-making. It is for this reason that the Statistics Division (DIVEST) proposes a model for forecasting the number of complaints for crimes that are registered in the Peruvian National Police stations in Metropolitan Lima, based on the RANDOM FOREST regression technique. REGRESOR AND REGRESSION DECISION TREE, with which it will be possible to know which of the two techniques generates a better prognosis based on the information that comes from SIDPOL and INEI.
In our country, the average number of complaints for crimes on a monthly basis in 2019 presents 30,000 cases, which has been increasing over the years, so it is of crucial importance to generate citizen security strategies that help to improve the well-being of Peruvian society, for this it is crucial to have reliable and quality information for decision-making; likewise, it has a Police Information and Communication Technologies Directorate (DIRTIC), which within its structure has in charge of the Statistics Division (DIVEST) and the Information Technology Division (DIVINFOR), whose function is to control the Police Complaints System in which 80 % of the Peruvian National Police stations are interconnected and there are 20 % that are not interconnected, for which reason 20 % of police stations do not record information in the Police Complaints System, generating uncertainty for decision-making. It is for this reason that the Statistics Division (DIVEST) proposes a model for forecasting the number of complaints for crimes that are registered in the Peruvian National Police stations in Metropolitan Lima, based on the RANDOM FOREST regression technique. REGRESOR AND REGRESSION DECISION TREE, with which it will be possible to know which of the two techniques generates a better prognosis based on the information that comes from SIDPOL and INEI.
Description
Keywords
Denuncia (Derecho penal)--Pronóstico--Modelos matématicos, Trámites gubernamentales--Perú--Automatización, Análisis de regresión
Citation
Endorsement
Review
Supplemented By
Referenced By
Creative Commons license
Except where otherwised noted, this item's license is described as info:eu-repo/semantics/openAccess