Machine learning–based yield prediction with IoT-enabled monitoring for sustainable Jowar cultivation

Jowar, an essential food grain grown in many regions is for helping to decide the food security. But its productivity is heavily affected by environmental factors: temperature, humidity, soil fertility and nutrient levels. This paper presents an innovative solution for improving Jowar crop yield using Machine Learning (ML) and Internet of Things (IoT) technologies. It is the main objective of this study to construct a predictive model, utilizing ML methods for prediction of Jowar crop yield. This model will rely on information of soil properties, environmental conditions (temperature, rainfall), nutrient concentration, latitude and soil ph. This model will help farmers to take crop management and resource allocation decisions factually leading to increase in the Jowar crop yield. ML is used to analyze and model the relationships with these data points. In the next stage, IoT will be implemented for Real time monitoring of Jowar crop using sensors-Environmental factors and soil parameters. Preliminary results show that the ML model can be used for estimating Jowar crop yield, and adoption of IoT technology would help to obtain meaningful on time data for aforementioned purpose. By integrating ML with IoT, the motivation of this research is to work toward sustainable improvement of Jowar crop for farmers to increase their productivity. The results indicate that the Gradient Boosting model effectively estimates yields of Jowar across Kharif and Rabi crop season used in the dataset. This is an example of how machine learning when incorporated with IoT-based monitoring could avert well-informed and green agriculture decision-making.