Implementación de algoritmo de machine learning supervisado para la predicción del comportamiento de las importaciones en Colombia de teléfonos celulares

Abstract

The implementation of machine learning algorithms has become a great advantage for decision making in any field. This is the reason why, a machine learning algorithm is used in this project while supervised in the programmation language Python in the execution of a forecast for the cellphones’ import in the country with the objective of studying its behaviour for the years 2020 and 2021; taking into account, as training data, the historical information contained in the DIAN’s webpage in the last ten years. To do that, we used a methodology that contemplated, as the first phase, the preparation of the data followed by the behaviour of the imports through graphics and charts; afterwards, we continued to the phase of modeling which was a key part since all the theoretical foundation for the validity of the model was concentrated on this stage. Finally, we proceeded with the evaluation of the model and then, the production.