Small Holders Farming Predictive Analysis Using Peer- To-Peer Approach

Abstract

Farmers cultivating crops without permission from agriculture centres and farmers might choose some crops to Grow without complete knowledge of climatic and soil conditions, resulting in crop failures and further losses due to low rates set by middlemen for selling the yield, leading to an increase in farmer suicides and a data scalability problem (i.e., every time details are updated). To address the aforementioned issues, the major focus is on establishing a suggested system that would provide farmers with crop recommendations and yield predictions based on soil and meteorological parameters such as pH, N, P, and K levels; temperature; rainfall; and humidity. With these circumstances, farmers can determine which crops are the most suited as well as profitable and how much yield can be created. Farmers may also sell their agricultural harvest directly to purchasers without the need for a third party. We may make the farmer register to start the crop, obtain acknowledgement from village centres, and then, after farming, re-enter the crop specifics and amount of crop grade to accomplish data scalability. The transaction has now been accomplished, from receiving the crop at this address to sending the harvest to a certain warehouse. In this, we can use supervised machine learning approaches like classification and regression to help farmers decide the best crop to grow to avoid crop failures due to climate change, as well as to eliminate the need for a third party to sell to buyers, resulting in direct profits to farmers and lowering farmer suicide rates.