A Bayesian Probability Model Can Simulate the Knowledge of Soybean Rust Researchers to Optimize the Application of Fungicides

A Bayesian Probability Model Can Simulate the Knowledge of Soybean Rust Researchers to Optimize the Application of Fungicides

Gregory Vinícius Conor Figueiredo, Lucas Henrique Fantin, Marcelo Giovanetti Canteri, José Carlos Ferreira da Rocha, David de Souza Jaccoud Filho
Copyright: © 2019 |Pages: 15
DOI: 10.4018/IJAEIS.2019100103
Article PDF Download
Open access articles are freely available for download

Abstract

Asian rust is the main soybean disease in Brazil, causing up to 80% of yield reduction. The use of fungicides is the main form of control; however, due to farmer's concern with outbreaks many unnecessary applications are performed. The present study aims to verify the usefulness of a probability model to estimate the timing and the number of fungicides sprays required to control Asian soybean rust, using Bayesian networks and knowledge engineering. The model was developed through interviews with rust researchers and a literature review. The Bayesian network was constructed with the GeNIe 2.0 software. The validation process was performed by 42 farmers and 10 rust researchers, using 28 test cases. Among the 28 tested cases, generated by the system, the agreement with the model was 47.5% for the farmers and 89.3% for the rust researchers. In general, the farmers overestimate the number. The results showed that the Bayesian network has accurately represented the knowledge of the expert, and also could help the farmers to avoid the unnecessary applications.
Article Preview
Top

Methodology

The model was constructed in the following three steps: (I) modelling to diagnose the disease, determined by relevant variables to predict an ASR epidemic; (II) adaptation of the created model into a qualitative model to estimate fungicide application; and (III) construction of the quantitative model. All of the developmental stages were performed using GeNIe version 2.0 (Druzdzel, 1999).

Complete Article List

Search this Journal:
Reset
Volume 15: 1 Issue (2024): Forthcoming, Available for Pre-Order
Volume 14: 1 Issue (2023)
Volume 13: 2 Issues (2022): 1 Released, 1 Forthcoming
Volume 12: 4 Issues (2021)
Volume 11: 4 Issues (2020)
Volume 10: 4 Issues (2019)
Volume 9: 4 Issues (2018)
Volume 8: 4 Issues (2017)
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 2 Issues (2012)
Volume 2: 2 Issues (2011)
Volume 1: 2 Issues (2010)
View Complete Journal Contents Listing