Abstract
Potatoes stand as one of the most vital staple crops globally, providing essential nourishment and sustenance to millions of people worldwide. Their significance lies in their versatility, nutritional richness, and ability to thrive in diverse climates, making them crucial for global food security. However, accurately forecasting potato production is paramount for effective agricultural planning and ensuring an adequate food supply. In this research endeavour, we introduce a novel approach to enhance the precision of potato production forecasts using advanced machine learning techniques. Our methodology revolves around employing long short-term memory (LSTM) models, which are optimised through the innovative Balance Dynamic Biruni Earth Radius Optimization Algorithm (BDBER). This algorithm dynamically adjusts exploration and exploitation strategies, effectively navigating the solution space to optimise the parameters of the LSTM model. By harnessing the power of machine learning and algorithmic optimization, we aim to improve the accuracy of annual potato production forecasts. To evaluate the efficacy of our approach, we compare the performance of the optimised LSTM models with traditional machine learning algorithms. Various performance metrics are scrutinised, and statistical tests, including ANOVA and Wilcoxon signed rank tests, are conducted to bolster the credibility of our findings. Our analysis reveals that the LSTM models optimised by BDBER surpass alternative methods, exhibiting superior accuracy and stability in potato production forecasting. Notably, the root mean square error (RMSE) of 0.00899 and fitted time of 0.00449 underscore the robustness of our approach. This study represents a pivotal contribution to the advancement of agricultural forecasting techniques. By providing more accurate and reliable predictions, our methodology equips policymakers and stakeholders with invaluable insights for informed decision-making. Ultimately, our research endeavours to bolster global food security and promote sustainable agricultural practices.
