Modelo predictivo que analice cuales son las zonas con mayor índice de peligrosidad en la ciudad de Guayaquil y a partir de esta información predecir qué sectores en desarrollo tendrán el mismo nivel de delincuencia.

Abstract

This research project contains the design of a predictive model that analyzes the areas with the highest index of danger in the city of Guayaquil and with this information predict which developing sectors will have an increase in crime using data mining and machine learning algorithms. This modeling will be done with the use of programming tools that will be used to clean the data, process it and apply supervised learning algorithms like linear regression and ramdom forest to train the model and predict future values of crime in different areas of Guayaquil and be able to be visualized in a data visualization software. The whole process from data collection to the analysis of the results obtained is part of the stages of the KDD or Knowlede Discovery in Databases (KDD) data mining methodology. The objectives of the research project consist of collecting and analyzing variables that describe a sector as dangerous, the design of the predictive model itself and the assessment of the predictions obtained after the creation of the model. Finally, the project concludes with the visualization of the regression algorithms in a visualization software that presents the areas that may become dangerous in the future.