Cognitive Data Clustering for Industrial Applications using IoT

Abstract

High-performance data analytic tools are essential in the era of ubiquitous connectivity, Internet of Things (IoT) devices, and massive data sets. The Industrial Internet of Things (IIoT) is a subset of the IoT that applies the benefits of machine-to-machine communication to industrial settings. The basic challenge with big data mainly consists of computational cost, expensive monitoring of equipment status, fault detection and serious delays. All of these have contributed to the shift from the conventional to the intelligent manufacturing paradigm. Clustering is a useful statistical tool or as a standalone analysis to find interesting patterns in a dataset. Because of data management's significance to the IIoT, taxonomy has been proposed to categorize the basic data management features. The proposed technique makes use of the underlying framework to manage massive data sets. This paper presents Clustering of IoT -based Big Data [CIoT- BD] for tracking the dynamics of data management processes. This aids in identifying and summarizing the big data tools and techniques used in IIoT. Data redundancy can be reduced through the use of deep learning-based techniques named Pooling Method to extract pertinent information of each defect. The simulation results demonstrate the effectiveness and performance of the suggested method, which is on level terms with or even more precise and speeder than methods employing the entire dataset based on the clustering algorithm.