Framework for cost-effective analytical modelling for sensory data over cloud environment

Abstract

In order to offer sensory data as a service over the cloud, it is necessary to execute a cost-effective and yet precise data analytical logic within the sensing units. However, it is quite questionable as such forms of analytical operation are quite resource dependent which cannot be offered by the resource constraint sensory units. Therefore, the proposed paper introduces a novel approach of performing cost-effective data analytical method in order to extract knowledge from big data over the cloud. The proposed study uses a novel concept of the frequent pattern along with a tree-based approach in order to develop an analytical model for carrying out the mining operation in the large-scale sensor deployment over the cloud environment. Using a simulation-based approach over the mathematical model, the proposed model exhibit reduced mining duration, controlled energy dissipation, and highly optimized memory demands for all the resource constraint nodes.