The use of artificial neural networks in analysing the nutritional ecology of Chrysomya megacephala (F.) (Diptera: Calliphoridae), compared with a statistical model

Artificial neural networks (ANNs) have been widely applied to the resolution of complex biological problems. An important feature of neural models is that their implementation is not precluded by the theoretical distribution shape of the data used. Frequently, the performance of ANNs over linear or non-linear regression-based statistical methods is deemed to be significantly superior if suitable sample sizes are provided, especially in multidimensional and non-linear processes. The current work was aimed at utilising three well-known neural network methods in order to evaluate whether these models would be able to provide more accurate outcomes in relation to a conventional regression method in pupal weight predictions of Chrysomya megacephala, a species of blowfly (Diptera: Calliphoridae), using larval density (i.e. the initial number of larvae), amount of available food and pupal size as input data. It was possible to notice that the neural networks yielded more accurate performances in comparison with the statistical model (multiple regression). Assessing the three types of networks utilised (Multi-layer Perceptron, Radial Basis Function and Generalised Regression Neural Network), no considerable differences between these models were detected. The superiority of these neural models over a classical statistical method represents an important fact, because more accurate models may clarify several intricate aspects concerning the nutritional ecology of blowflies.

Perda de matéria seca em grãos de milho submetidos a sistemas de secagem natural e artificial.

The permanence of the corn grain in the field, after physiological maturity, is an important cause of crop losses, both in quantitative and qualitative aspect. By ceasing the supply of assimilated substances to grains, due to physiological maturity, the synthesis reactions are overcome by breathing, responsible for the maintenance of the living tissues of the grains, which occur at the expense of reserves accumulated during grain formation. In addition, there are losses from fungus and insects attack due to adverse weather conditions. Technological advances in recent decades, the develop of grain dryers with different capacities and efficiencies, has led to the early withdrawal of the product from the field, still damp, reducing spoilage. Moreover, the use of artificial drying systems can represent a significant cost to the producer. Thus, the present work aimed to study the effect of natural and artificial methods of drying on maize dry matter losses, for Botucatu, city of Sao Paulo state, Brazil. The cornfield production was conducted at the Experimental Farm “Lageado” and the experimental treatments were conducted in the Laboratory of Agricultural Products Processing, in the Department of Rural Engineering, where the drying systems were tested. The treatments were: shade (control), artificial with heated air, artificial unheated air and, drying attached to the plant. The following analyzes for quality monitoring were performed: weight test, thousand-grain weight test and, grain dry weight. The results showed significant loss in quality of drying beans attached to the plant, by assessing the dry matter loss and by the variation of the grain weight. The weight test showed that the worst performance was the artificial with heated air treatment. We used mathematical modeling techniques to describe the dry matter loss and adjusted the mathematical model to the experimental data analyzed. From the experimental data obtained during drying the grain attached to the plant, it was still possible to fit a regression model that estimates the loss of grain dry matter under the climate from Botucatu during the 2011/2012 harvest period.