An IoT-based soil and plant nutrient management of papaya production in Bangladesh

Abstract

This article gives an insight into the creative use of the IoT system in the case of the regulation of the soil and the plant nutrients in the cultivation of papaya plants in the farming landscape of Bangladesh. Papaya production in Bangladesh faces challenges related to soil quality, nutrient management, and the environment. In this study, we propose an IoT-based system for soil and plant nutrient management to improve papaya production. This paper describes implementing an IoTenabled soil and plant nutrient management gadget centered on papaya production in Bangladesh. This study develops superior predictive models in ARIMA, SARIMAX, Random forest, XGBoost, Linear Regression, Logistic Regression, and Multilinear Regression to optimize the rural activities by properly estimating soil and plant nutrient stages. The proposed framework consists of an IoTsensor network around the field in strategic places across the papaya fields, monitoring and recording moisture degrees, Nitrogen (N), Phosphorus (P), and Potassium (k) tiers, amongst different required degrees. The records assets heterogeneously encompass the moisture stage information and information on the essential nutrients, namely Nitrogen (N), Phosphorus (P), and Potassium (k), which are statistically modeled with environmental parameters to expect the future nutrient degrees and are expecting the general health of the papaya crop. This information is then fed into the advanced predictive fashions that make up the superior framework—ARIMA, SARIMAX,Random forest, XGBoost , Linear Regression, Logistic Regression, and Multilinear Regression.In correlating the models, one-of-a-kind sets of model assessment metrics are used. suggest Squared blunders (MSE) and (RMSE) degree the value of average squared variance or the unfold between determined and expected values. Accuracy metrics determine how correct the predictions are, with special attention given to binary classification like nutrient deficiency. suggest Absolute Scaled blunders (MAPE) is a normalized measure of forecast accuracy that allows scale evaluation. Consequences will generate the website precise, dynamic, and really useful information that may be implemented on actual grounds via the right choice-making to decorate nutrient management alternatives, accordingly enhancing the performance and sustainability of papaya manufacturing.