Student Poster Display
Plant-Insect Ecosystems
Student
Student Competition
Kayo Heberth de Brito Reis
University of Nebraska
North Platte, Nebraska
Alisson Santana (he/him/his)
Post-Doctoral Research Associate
University of Nebraska
North Platte, Nebraska
Jhersyca Paes
Federal University of Viçosa
Viçosa, Minas Gerais, Brazil
Marcelo Coutinho Picanço
Federal University of Viçosa
Viçosa, Minas Gerais, Brazil
Julie Ann Peterson
Professor
University of Nebraska
North Platte, Nebraska
Renato Sarmento
Federal University of Tocantins
Gurupi, Tocantins, Brazil
Striacosta albicosta is an important pest that feeds mainly on corn, causing significant economic damage to US production. Artificial Neural Networks (ANNs) are computational systems that simulate the functioning of a biological neuron and have several applications in pest population modeling, including predicting insect population density. Thus, our study aims to develop an ANN capable of determining the population density of S. albicosta in corn crops and identifying the environmental factors related to its population outbreaks. The pest population density in the field and meteorological data were used to create the ANN using the software Rstudio. The ANN has the following topology: an input layer with the predictors, which are the environmental variables; a hidden layer with several neurons determined based on the dataset obtained; and an output layer, which is the population density of S. albicosta. The neurons in each layer were connected through synapses, and their weights were calculated through supervised machine learning based on the intended result. Several ANNs were created, and the most accurate one was chosen based on defined criteria. Our objective is to select an ANN that is capable of predicting the population density of S. albicosta in corn crops and determining the environmental factors correlated with population outbreaks of the pest.