ABSTRACT
Many economies, especially those in rising nations, are based on agriculture. The proper crop needs to be planted in the right location for the best outcomes. The majority of farmers and other agricultural activists promote crops using shoddy, tried-and-true scientific methods, which is a concern. In order to boost productivity and profit from the suggested technique, our proposed endeavor would help farmers choose the best crop based on elements like cost of cultivation, cost of production, and yield. We attempted to build an Internet of Things (IoT)-based crop recommendation system in this study. The user must enter parameters in order to obtain the data required to recommend the optimum crop. A set of data containing six various characteristics, such as humidity, temperature, nitrogen, phosphorus, and potassium content, needed to grow particular crops. Details about 21 various crops that could grow there were supplied. The dataset's inclusion of input parameters and their output (the type of crop) made it a classification challenge that required the application of a supervised machine learning method. K-Nearest Neighbor (KNN) was the algorithm of choice to calculate the distance. Our research produced a website that, when a user enters their soil's specifics, he or she uses a machine learning model to suggest the ideal crop.
