Bibliographic citations
Sueldo, B., Peña, E. (2022). Servicio web para la predicción de los delitos de robo utilizando machine learning en la ciudad de Lima Metropolitana [Tesis, Universidad Peruana de Ciencias Aplicadas (UPC)]. http://hdl.handle.net/10757/660716
Sueldo, B., Peña, E. Servicio web para la predicción de los delitos de robo utilizando machine learning en la ciudad de Lima Metropolitana [Tesis]. PE: Universidad Peruana de Ciencias Aplicadas (UPC); 2022. http://hdl.handle.net/10757/660716
@misc{renati/400130,
title = "Servicio web para la predicción de los delitos de robo utilizando machine learning en la ciudad de Lima Metropolitana",
author = "Peña Cueva, Edwin Daniel",
publisher = "Universidad Peruana de Ciencias Aplicadas (UPC)",
year = "2022"
}
The objective of this thesis work is to implement a proposal for the development of a technological solution that allows the prediction of robbery crimes in the city of Lima using artificial intelligence. In the proposed solution, the exposure of API web services has been proposed, which may be consumed with security tokens by web platforms or cell phone applications, through these web services users will be able to register crimes and consult the predicted areas of theft. by means of coordinates. Likewise, within the proposal the use of an artificial intelligence model based on machine learning for the prediction of crimes is proposed, the proposed solution and the predictive model are deployed in the Azure cloud. The contribution of the project stands out for its innovation by using a predictive algorithm of artificial intelligence, designing an architecture oriented to microservices and having an infrastructure in the Azure cloud, in this way it seeks to guarantee optimal performance in the solution and have a scalable infrastructure that supports a high concurrence of users. On the other hand, the proposed software seeks to provide information to citizens so that they take prevention based on the knowledge of future crimes, it is also about helping the National Police of Peru to take a more proactive stance by knowing geographic locations where crimes will occur. to future. Different cloud platforms have been analyzed that allow the design of the predictive model of theft crimes, where the selection criterion was if the platform provided machine learning services, the costs of the service, a free trial for concept tests, ease of use. , and also check if it provides a repository of example artificial intelligence experiments. In addition, an analysis and compilation of information has been carried out through a survey to find out the opinion of the public regarding good practices that must be considered in order to have an adequate process for registering and dealing with complaints, based on which a under the maturity level of the current process where it is evident that there are points to improve such as the information provided to citizens about crimes, the crime registry, the application architecture and infrastructure, etc. These points are considered and improved with the proposed solution.
This item is licensed under a Creative Commons License