Bibliographic citations
This is an automatically generated citacion. Modify it if you see fit
Escobedo, M., Tapia, C. (2024). Comparativa de modelos de regresión a fin de predecir el crimen en zonas de alto riesgo de la ciudad de Lima [Universidad de Lima]. https://hdl.handle.net/20.500.12724/21714
Escobedo, M., Tapia, C. Comparativa de modelos de regresión a fin de predecir el crimen en zonas de alto riesgo de la ciudad de Lima []. PE: Universidad de Lima; 2024. https://hdl.handle.net/20.500.12724/21714
@misc{renati/1439724,
title = "Comparativa de modelos de regresión a fin de predecir el crimen en zonas de alto riesgo de la ciudad de Lima",
author = "Tapia Aquino, Cynthia Lizet",
publisher = "Universidad de Lima",
year = "2024"
}
Title: Comparativa de modelos de regresión a fin de predecir el crimen en zonas de alto riesgo de la ciudad de Lima
Authors(s): Escobedo Neyra, María Cielo; Tapia Aquino, Cynthia Lizet
Advisor(s): Gutiérrez Cárdenas, Juan Manuel
OCDE field: https://purl.org/pe-repo/ocde/ford#2.02.04
Issue Date: 2024
Institution: Universidad de Lima
Abstract: La delincuencia sigue siendo un problema en Lima Metropolitana, Perú, que afecta a la sociedad. Este artículo tiene como objetivo analizar los delitos contra la propiedad y reconocer la falta de estudios para predecir estos crímenes. Para solucionar este problema, se utilizan técnicas de regresión como Extra Tree, XGBoost, Bag, AdaBoost, Support Vector y Random Forest. Mediante GridSearchCV se optimizan los hiperparámetros para mejorar los resultados de la investigación. El modelo de Extra Tree Regression muestra un coeficiente de determinación (R2) de 0,79, y se evalúan métricas de error como el error cuadrático medio de la raíz (MSE), el error cuadrático medio (RMSE) y el error absoluto medio (MAE). Este enfoque considera patrones temporales de delincuencia para resolver la incertidumbre y combatir la inseguridad ciudadana.
In Metropolitan Lima, Peru, crime is still an issue that has an impact on society. The purpose of this paper is to examine property crimes and acknowledge the paucity of research on their predictiveness. Regression approaches including Extra Tree, XGBoost, Bag, AdaBoost, Support Vector, and Random Forest are utilized to overcome this issue. The study outcomes are enhanced by optimizing the hyperparameters with GridSearchCV. Error measures including mean absolute error (MAE), root mean square error (RMSE), and root mean square error (MSE) are assessed for the Extra Tree Regression model, which has a coefficient of determination (R2) of 0.79. This strategy resolves ambiguity and counteracts citizen uneasiness by taking temporal trends of crime into account.
In Metropolitan Lima, Peru, crime is still an issue that has an impact on society. The purpose of this paper is to examine property crimes and acknowledge the paucity of research on their predictiveness. Regression approaches including Extra Tree, XGBoost, Bag, AdaBoost, Support Vector, and Random Forest are utilized to overcome this issue. The study outcomes are enhanced by optimizing the hyperparameters with GridSearchCV. Error measures including mean absolute error (MAE), root mean square error (RMSE), and root mean square error (MSE) are assessed for the Extra Tree Regression model, which has a coefficient of determination (R2) of 0.79. This strategy resolves ambiguity and counteracts citizen uneasiness by taking temporal trends of crime into account.
Link to repository: https://hdl.handle.net/20.500.12724/21714
Discipline: Ingeniería de Sistemas
Grade or title grantor: Universidad de Lima. Facultad de Ingeniería
Grade or title: Ingeniero de Sistemas
Juror: Nina Hanco, Hernán; Irey Núñez, Jorge Luis; Ramírez Cerna, Lourdes
Register date: 12-Dec-2024
This item is licensed under a Creative Commons License