eKisaan -A B2B Agricultural Platform to Enhance the Production of Crops using Machine Learning

Agriculture is the industry that deals with growing crops and raising livestock. Modern agronomy, plant breeding, agrochemicals like chemical fertilizers, and technological advancements have all contributed to a significant increase in crop yields, but they also have negative effects on the ecosystem and the environment. The effects of agriculture include antibiotic resistance, deforestation, aquifer depletion, global warming, and other environmental problems. One of the main challenges in agriculture is that a crop’s price has dropped because it was not grown at the right time and the right season. Due to lack of right information and fluctuating weather conditions, there’s a need to develop a system where the farmers can be provided with the information on the types of crops to be grown or nurtured. The main aim of our system is to help farmers to make appropriate decisions regarding the crop and fertilizer selection in order to increase the crop’s pro ductivity. The proposed system performs analysis based on the geographical location and other factors such as weather, temperature, humidity, rainfall, phosphorus and nitrogen levels using algorithm viz. Random Forest, Decision Tree and Naive Bayes. By performing exploratory data analysis through these algorithms, it was found that the Random Forest algorithm yields the most accurate and unbiased prediction. In order to connect farmers directly with the retailers for trading of crops, the proposed system provides the B2B platform. This eliminates any third party involvement and thus any transaction fee.