Intelligent Modeling of Sugar-Cane Maturation

The text area is 29 × 45 picas (~12.3 cm × 19 cm) to fit in the printed proceedings book without reduction. When printed on letter paper there will be wide margins. Abstract. Previous use of Artificial Intelligence (AI) in agriculture for forecasting productivity indicators, especially Artificial Neural Networks (ANN), has shown that it is possible to approximate sugarcane maturation curves. However, ANNs are widely known to offer some difficulties to be parameterized; normally, some heuristics and devotion by the users are necessary to provide a suitable parametrical selection. In this work we propose a tool that searches ANN parameters automatically (i.e. without direct interference of users in this task). For this high goal, we utilized another artificial intelligent technique: Genetic Algorithms (GA). The paper concludes showing results of ANNs which parameters have been provided by (the contributed) automated approach; comparisons to manually setup ANNs are included.