Implementación del algoritmo de agrupación gravitacional para determinar un patrón de conocimiento delictivo y la relación existente entre reincidencia y el consumo de sustancias psicoactivas en jóvenes infractores de la ley

Abstract

The purpose of this project was to implement the algorithm of gravitational grouping, in young offenders of the law to find a criminal pattern and the relationship between consumption and recidivism. To this end, the data collection was carried out in the Amigoniano CIPA educational center, through the REFORMA system. Data were collected from 746 young people between 14 and 18 years old. Before the analysis, the data went through the process of detection and correction, treating missing data, selecting attributes and carrying out the digitalisation of the data given that the algorithm handles only data of numerical type. The algorithm takes each data as an object in the future space and compares them with a random object, moving them by means of gravitational force and the second law of newton movement, if it is determined that the distance after the movement is less than epsilon, the objects make up a group, the groups are stored in a disjoint set, the algorithm automatically determines the number of groups, returning in a flat file the created groups, for the analysis and extraction of knowledge. The algorithm was compiled in java and the algorithmic complexity was calculated.