An Optimal Algorithm to Resolve the Impact of Big Data in IoT Application using Bayesian Network

Abstract Internet of Things (IoT) becomes the hot cake in all technological fields. IoT applications concurrently generate the huge amount of data that need to be handled. In this paper, an optimal mechanism is proposed to handle big data for decision making. Based on the classical decision-making process in the field of Big Data, we design an adaptive knowledge-based method through IoT application. A Bayesian network model is used to manage knowledge formation in all direction for the decision-making process. Knowledge of Bayesian networks is habitually emitted as an optimal solution, where the analysis job is to locate a formation that exploits a statistically motivated score. Generally, accessible knowledge tools deal with this optimal solution by means of ordinary search methods. As it required big amount of data space, therefore it is a time consuming procedure that should be avoided. The situation becomes decisive once big data involve in searching for optimal solution. An algorithm is introduced to accomplish faster processing of optimal solution by limiting the search data space. The proposed recursive algorithm will limit the search space. The result shows that the optimal mechanism of the decision process is able to handle big data by reducing processing time, computational complexity and a higher percentage of prediction rates.