An Efficient Automated Hybrid Algorithm to Predict Floods in Cloud Environment

Abstract

Natural and environmental sciences are one of the scientific domains which seek a lot of attention as it requires accurate real time predictions. In particular, flooding induced by heavy precipitation is one of the regular risks in Eastern Indian states. In this research work, the state of Odisha, India have been selected for predicting floods because majority of the state’s districts have been exposed to floods, leading to unprecedented loss of life and property. In this paper, an optimization based feature selection Genetic Algorithm (GA) have been combined with classification algorithms to predict the occurrence of floods. The experimental results show that the GA-SVM algorithm outperforms in terms of accuracy and total execution time in comparison to other hybrid algorithms. Finally, the results are validated and compared by executing the proposed hybrid algorithm over the heterogeneous resources in Cloud environment.