Rice Crop Yield Prediction using Machine Learning and Integrating IoT

Abstract: This paper focuses on predicting the rice harvest and investigating the factors affecting rice production in various regions of the Maharashtra region of India. The software aims to provide a rice harvest using a random forest algorithm and accurately predicting the yield. To demonstrate the effectiveness of harvest forecasting, an Indian government database will be used in 34 districts of the Maharashtra region, India [2]. Boundaries such as rainfall, temperature, humidity, and location are given as a contribution to the random forest model to define the annual variation of the regional rice crop in Maharashtra. The software will also use other IoT devices to retrieve real-time data from the field. This will give an accurate result to farmers and prevent major losses to farming. With the help of powerful services like Amazon Web Services(AWS), Java, and Flask it tends to work on low-end devices and remote regions.

Keywords: Amazon Web Services(AWS), Flask, REST API, Android Studio, MySQL, Java, Arduino Ide and Random Forest Regression.