Machine Learning Facilitated Rice Prediction in Bangladesh

The climate of a region is often determined by its landscape and amount of vegetation present in it. Environment parameters such as rainfall, wind-speed and humidity are highly influenced by these alluvial features. Bangladesh, a country situated on the banks of the Himalaya, does not have a homogeneous topography. Human settlement over the course of centuries has led to pockets of micro-regions. Each of those regions has a different micro climate. An entrepreneur involved in the food industry therefore has to carefully choose regions of land that will give him/her the desirable production. In this study a research initiative has been taken to predict the yield of crops using machine learning models. The models were at first trained on the correlation between past environmental patterns and crop production rate. Then the models are compared to measure their effectiveness on unknown climatic variables.