Optimizing Artificial Neural Network-Based Models to Predict Rice Blast Epidemics in Korea

Abstract

To predict rice blast, many machine learning methods have been proposed. As the quality and quantity of input data are essential for machine learning techniques, this study develops three artificial neural network (ANN)-based rice blast prediction models by combining two ANN models, the feed-forward neural network (FFNN) and long short-term memory (LSTM), with diverse input datasets, and compares their performance. The Blast_Weather_FFNN model had the highest recall score (66.3%) for rice blast prediction. This model requires two types of input data: blast occurrence data for the last 3 years and weather data (daily maximum temperature, relative humidity, and precipitation) between January and July of the prediction year. This study showed that the performance of an ANN-based disease prediction model was improved by applying suitable machine learning techniques together with the optimization of hyperparameter tuning involving input data. Moreover, we highlight the importance of the systematic collection of long-term disease data.

Keywords: artificial intelligence; artificial neural network; machine learning; rice blast.

Fig. 1

Distribution map of National Crop Pest Management System (NCPMS) data and weather observation data used in the study. Blue dots indicate the NCPMS data of rice blast occurrence and red dots indicate the weather station sites for weather observation data.

Fig. 2

A flowchart of the development of Blast_Weather_FFNN and Blast_Weather_LSTM models used in this study. Blast_FFNN does not include weather data indicated as the light green box in the figure. FFNN, feed-forward neural network; LSTM, long short-term memory; NCPMS, National Crop Pest Management System.