Spatio-Temporal Clustering for Environmental Data: India

Data mining is the mathematical approach of exploring new information by assessing definitely known facts and figures. The paper applies this approach to the real-world environmental data The Greenhouse Gas Emissions for the country India. The greenhouse gas effect is a gradual phenomenon of increase in the average temperature of the earth's atmosphere due to the raised emissions and concentrations of greenhouse gases. Global Warming, which is a topical issue, has led to extreme global climate change posing as a mammoth threat to the foods security situation in India with recurring and severe droughts and ravaging floods engulfing the arable land. In this paper, a novel unsupervised learning phenomenon has been used in finding out the high and low Greenhouse gas emissions? sectors in India comparing with China and Indonesia (top three highly populated Asian countries) for fair analysis. The three countries (incorporating space) and their consecutive 15 years data (incorporating time) result into large spatio-temporal data sets that has raised the need of performing spatio-temporal clustering on the data sets. The CRISP-DM methodology, K-means clustering algorithm and Weka tool has been used to design and develop the model which can be used to analyze environmental data by governmental authorities when decisions on such data are to be made and will also provide deeper in-sight of the data, thus, contributing towards the widespread efforts to reduce the greenhouse gas emissions to mitigate weather changes and promote cleaner energy sources.