ANALYSIS OF THE CAUSES OF DRUG USE BY MEANS OF MACHINE LEARNING

Drug consumption is a real problem that affects a large part of the world's adult society, but unquestionably, young people are a very vulnerable sector within this scenario, a situation from which Ecuador is not exempt. Due to the need to study this, the main objective of this research was to analyze the causes of drug use and help prevention through supervised or unsupervised automatic learning techniques. Additionally, the project's objective was to provide techniques and tools on automatic learning and its practical application in related problems. A sample of 3,876 Ecuadorian youths and adolescents was analyzed as a case study and the results were processed using Orange software, and a risk of 85% was obtained. The tasks defined in the CRISP DM methodology were executed without novelties, and it was possible to indicate that the data are sufficient for modeling, and that the predictions made are reliable.