Data Analytic Techniques for Developing Decision Support System on Agrometeorological Parameters for Farmers

The day-to-day lives of humans are changing remarkably due to the evolution in tools, techniques and technology across the planet. This evolution is not only impacting the growth of humans but also contributing to the growth and well-being of society and country. The domain of data analytics (DA) and internet of things (IoT) is very much facilitating this growth. But there have been only a handful of innovations and explorations in the field of agriculture, although it being the backbone and largely contributing to the gross domestic product (GDP) of a country like India. The reason for it may be profuse, such as the erratic weather conditions, improper irrigation, farmers being skeptical using modern tools and many more. But being in a developing country that has its primary focus on invention and innovation, a consensus has to be reached so that the modern tools and technologies, abet agriculture throughout the country. In our work, an attempt is made to analyze the different aspects that influences the variable outcomes in agriculture with the aid of various data analytic algorithms. Rainfall, humidity and temperature are some of the variables that determine the type of crop. Therefore, the task of prediction of crop type given these factors using decision trees and support vector machines (SVM) is implemented, and the accuracy of the models are computed. Here, more focus is given to the state of Karnataka and to its major crops. With rice, ragi and maize being some of the predominant crops, an analysis is portrayed considering the yield across the state.