Machine Learning Technique for Prognosticating Crop Yield Variability with Air Quality Index

Rice production in India holds significant importance in terms of its contribution to the economy and food security. However, the expansion of this sector is currently facing challenges due to the increasing levels of air pollution, including ground-level ozone and particle matter. The objective of this study is to examine the influence of air quality index (AQI) factors on rice productivity, namely during the winter season, with a particular emphasis on the exceptional year of 2020-2021. By utilizing Internet of Things (IoT) technology, the collection of real-time air quality data is combined with rice yield analysis through the implementation of decision tree regression, k-nearest neighbors (KNN), and support vector regression (SVR) methods. The research findings emphasis the effectiveness of the decision tree model in accurately representing the complex relationship between AQI and rice yield. This model demonstrates higher predictive accuracy when compared to SVR and KNN. This study highlights the importance of managing air quality in order to achieve sustainable rice production. Additionally, it emphasizes the potential of monitoring systems based on the Internet of Things (IoT) to significantly transform agricultural practices. By doing so, these systems may effectively connect environmental stewardship, technological innovation, and food security.