부산대학교 도서관

Time Series Analysis plays a crucial role in comprehending and predicting trends across various domains, including agriculture. This study focuses on analyzing historical data related to the sales of sowing crops and employs machine learning techniques to make data-driven predictions. The primary objective is to assist farmers, policymakers, and stakeholders in optimizing crop production and enhancing agricultural sustainability. This research integrates historical sales data of sowing crops with meteorological and socio-economic factors to construct robust predictive models. Various machine learning methods, such as linear regression, Random Forest, and Decision Tree, are implemented and compared to identify the most suitable model for precise forecasting. Additionally, seasonal patterns, cyclical trends, and external factors like climate conditions and market demands are considered to improve the model’s accuracy. The findings of this study have significant implications for both small-scale and large-scale farming operations. By harnessing the potential of machine learning, farmers can make well-informed decisions regarding planting and harvesting, resource allocation, and risk management in crop sales. Policymakers can utilize these insights to design more effective agricultural policies that support sustainable crop production and rural livelihoods. In conclusion, this research highlights the value of time series analysis and machine learning in the agricultural sector, providing a valuable tool for enhancing crop sales forecasting. The integration of datadriven insights into farming processes can lead to increased productivity, reduced waste, and a more resilient agricultural industry in the face of changing environmental and market conditions