Title: Performance analysis of hybrid classification system model for big data stream using internet of things

Authors: N. Gayathri Devi; K. Manikandan

Addresses: School of Computer Science and Engineering, Vellore Institute of Technology, Vellore – 632014, Tamil Nadu, India ' School of Computer Science and Engineering, Vellore Institute of Technology, Vellore – 632014, Tamil Nadu, India

Abstract: Big data and internet of things (IoT) have become the world's prominent technology and reached a high impact on millions of people's daily life that helps in managing environmental and physical systems processed in real-time. In this research, we proposed a hybrid classification system model named multi-output regression with deep belief networks (MOR-DBN) to improve the performance of the classifier when huge amount of streaming data is transferred from IoT devices. Moreover, the improved privacy-preserving rotation based condensation algorithm (P²RoCAl) with Geometric Transformation is also used for obtaining high utility for data streaming to protect from various kinds of attacks during data reconstruction. The simulation results show that proposed framework obtains a high accuracy when compared with other existing algorithms in terms of precision, recall, and F-measure. Overall the proposed MOR_DBN model outperforms by obtaining a 96.21% for SSDS dataset, 97.89% for FRDS dataset, 95.7% for HPDS dataset, and 99.23% for ESDS dataset.

Keywords: MOR; multi-output regression; DBN; deep belief network; P²RoCAl; privacy preserving rotation based condensation algorithm; geometric transformation; data streaming; big data; IoT; internet of things.

DOI: 10.1504/IJSSE.2021.121438

International Journal of System of Systems Engineering, 2021 Vol.11 No.3/4, pp.336 - 351

Received: 21 Jul 2020
Accepted: 09 Nov 2020
Published online: 14 Mar 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article