Estudio de aplicabilidad de técnicas de inteligencia artificial en el sector agropecuario

Abstract

Machine learning is a branch of artificial intelligence that uses algorithms to perform tasks, without having been programmed explicitly. For its operation requires a process of training and validation based on examples. In this thesis the application of artificial intelligence techniques in agricultural production is studied. As main result of the thesis, three articles has been published in journals with important JCR impact factors. Two of them refer to a database of poultry production of eggs and the other to a database of the industrialization of sugar cane. In poultry production these techniques were studied for the early warning of problems in the production curve. For the application of these techniques in the industrial process of sugarcane, the calibration models of the NIR spectra for the quality control in a sugar factory were optimized. In this work were used Support Vector Machines and Artificial Neural Networks. The application of these techniques has a high potential of use in the agricultural production, since it opens up the development of intelligent systems to support productive decisions.