IOT Based Water Management System For Crops Using Conventional Machine Learning Techniques

Agriculture is one of the main sources of living for a large population in India, and plays a crucial part in the improvement of food production. “Smart farming” is an emerging technique employing evolving technologies like the Internet of Things (Io‘I’), the main thrust with expanded agrarian creation at an affordable cost, to attain more productivity in a sustainable path. With the assistance of IoT inclusive of Machine Learning, we can increase the efficiency of crop production. The fundamental point of this work is to build a system which can monitor and assist crop production, with minimum water utilization. India has 18% of the total populace, having 4% of the world's freshwater, out of which 80% is utilized in farming. Simultaneously, today many farmers hit by lower yield due to the dry climatic condition, are battling to remain alive. To settle this condition, effective administration of water ought to be created. In this work, a system is proposed, that can assist farmers in crop management by getting the data related to crop management viz. temperature, soil moisture and humidity. In this paper, the analysis of crops such as coriander and fenugreek with the help of IoT and supervised learning classifiers such as K-Nearest Neighbour, Random Forest, Decision Tree and Support Vector Machine is done. By using these Machine Learning techniques, the crops are classified as healthy and unhealthy category, to arrive at a decision whether to water them or not and subsequently, a motor is instructed to operate. The results of the study showed that out of all the four predicted algorithms the Binary SVM classifier algorithm showed the best results which is further implemented in the model.