Discretized Support Vector Prediction Classifier for Big Data Computation and Information Sharing in Cloud

The major challenges in Big Data applications are to explore the huge volumes of data and extract valuable information or knowledge for future measures. The distributed data mining on huge amount of Cloud data require minimal computational overhead and communication costs. However Classification performance of information being shared is not effective. In order to overcome such limitations, Discritized Support Vector Classification and Prediction (DSV-CP) model is proposed in this paper. The main goal of DSV-CP model is to provide efficient Big Data computation and information sharing in Cloud computing environment. Initially, preprocessing based on Interval Equivalence Discretization is performed to remove the unwanted and noise data obtained from various sources. This in turn reduces the computation time and space complexity involved during information sharing in Cloud environment. Next, DSV-CP model uses Support Vector Prediction classifier to classify the data based on user query request using parallel hyperplane aiming at improving the classification accuracy (i.e. search accuracy of user request information on big data). Finally, DSV-CP model accurately identify, the user requested information on big data with the help of classified data. Experimental evaluation of DSV-CP model is done with the performance metrics such as computation time, space complexity, classification accuracy and prediction rate. Experimental analysis shows that the DSV-CP model improves the classification accuracy by 12% and reduces the space complexity by 8% when compared to the state-of-the-art works.