Novel Hybrid ARIMA-BiLSTM Model for Forecasting of Rice Blast Disease Outbreaks for Sustainable Rice Production

DOI: https://doi.org/10.21203/rs.3.rs-2103058/v1

Abstract

In recent years, the application of artificial intelligence (AI) in agriculture has grown to be the most important research domain. The proposed work focuses on forecasting of rice blast disease outbreaks in paddy crop. Disease management in the farm fields is the most difficult problem on the planet. There are variety of reasons for this, first lack of farmers experience in diagnosing diseases, second experts experience in detecting diseases visually, third unfavourable climate. Recent days, researchers have offered variety of time series techniques in different applications. This study adds time series techniques to the field of agriculture by forecasting crucial rice blast disease outbreaks in paddy crop of Davangere region based on daily weather data obtained from KSNDMC. The statistical time series technique called ARIMA is trained by employing real data of blast disease outbreaks in Davangere region from the period of 2015–2019. Meanwhile deep BiLSTM model is trained by employing real weather data and blast disease outbreaks of Davangere region. Both the models are evaluated by performance metrics such as mean squared error and mean absolute error. The proposed research is focused on hybrid model ARIMA-BiLSTM which is a combination of statistical ARIMA model and deep BiLSTM model. Seasonal component of rice blast disease outbreak feature is extracted from additive decompose function used in ARIMA model and fed as dependent feature for BiLSTM model. According to the results obtained, the hybrid approach has the ability to successfully forecast blast disease outbreaks in paddy crop with mean squared error 0.037 and mean absolute error 0.028 compared to statistical ARIMA and deep BiLSTM model.

Full Text

This preprint is available for download as a PDF.