Environment Monitoring System for Agricultural Application using IoT and Predicting Crop Yield using Various Data Mining Techniques

Nowadays, Internet is facing an exponential increase in the number of electronic devices interfaced through it. Previously, there is a limit of connecting only mobile and PCs (personal computer) with the internet but now millions of devices can be connected using the concept of IoT (internet of things). Hence, IoT is able to communicate data between machine to machine and the data that is previously available to private server now easily available to the internet so that it can be accessed efficiently. This paper show the increase in the usage of IoT in monitoring like applications. Smart farming is the technique which intend to provide all the necessary resources for the specified amount of time. These resources required are light intensity, ambient temperature required, relative humidity, soil moisture content, pH reading of the soil. The central idea is to sense all these parameters one by one and take the final decision accordingly. A sensor node should be developed for sensing all the required resources parameters and subsequently send the data to the cloud for further processing. After getting all the required resources values our final aim is to predict the crop production using various data mining techniques. As the human lives are primarily depended on the food resources, the agricultural process need to be efficient and this efficiency can be enhanced if there is an accurate amount of resources. By getting the accurate amount of resources, Data mining techniques such as Random forest, KNN, SVM are used to analyze the crop production in advance such that farmer always have an upper hand on it and by comparing with the previous trend he will be able to detect which kind of parameter is accurate and which is not. A User defined site is designed to monitor all these values of agricultural process.