Hybrid Optimization Algorithm for Providing Big Data Classification

Abstract

The larger amount of databases is required classification for providing efficiency in big data applications. The parallel computing approach utilizes the classification whenever the uncertainty occurs in big data. The information entropy causes through the partitioning issues while compressing the data to a restricted amount of memory. The computational time is high in the process of iterations in to big data classification and the available division in the big data reduces the threshold level. The randomly assigned classification involves the identification of the optimal path for every attribute. The computation of information gain for every attribute will enhance the computation time. The optimization based scheduling will reduce the computation time. The hybrid optimization algorithm uses to obtain the optimal path with attribute based randomly assigned and scheduling process. The experimental results show that the proposed technique has the enhanced optimization compared with the other techniques.