Crop Yield Prediction Using Linear Regression and Random Forest Modelling

Global warming and climate change profoundly influence life on Earth, significantly impacting ecosystems. Changes in temperature, precipitation patterns, and frost timing affect water availability, crop yields, and the length of growing seasons. These variations may lead to region-specific changes in agricultural practices - enabling the cultivation of new crops in some areas while complicating agriculture in others. This study investigates the impacts of climate change on agricultural yield, focusing on staple cereal crops such as wheat, corn, millet, maize, linseed, olives, and mustard. Leveraging machine learning, this research develops a comparative model to predict crop yields in three key steps. Firstly, weather and crop yield data were collected from various countries over the past two decades. Secondly, data exploration and preprocessing were conducted to prepare datasets for analysis. Lastly, machine learning models were trained using two algorithms: Linear Regression and Random Forest. A comparative analysis of these models' accuracy is presented, with results validated through empirical evaluation.