Next Article in Journal
A Method for Fuzzy Soft Sets in Decision-Making Based on an Ideal Solution
Next Article in Special Issue
System Framework for Cardiovascular Disease Prediction Based on Big Data Technology
Previous Article in Journal
Factors Affecting the Perception of 3D Facial Symmetry from 2D Projections
Previous Article in Special Issue
Toward Bulk Synchronous Parallel-Based Machine Learning Techniques for Anomaly Detection in High-Speed Big Data Networks
Open AccessArticle

Self-Adaptive Pre-Processing Methodology for Big Data Stream Mining in Internet of Things Environmental Sensor Monitoring

1
Department of Computer and Information Science, University of Macau, Taipa 999078, Macau
2
School of Computer Science, North China University of Technology, Beijing 100144, China
3
Department of Computer Science, Electrical and Space Engineering, Lulea University of Technology, SE-97187 Lulea, Sweden
4
Department of Information Technology, Durban University of Technology, Ritson Campus, Durban P.O. BOX 1334, South Africa
*
Author to whom correspondence should be addressed.
Symmetry 2017, 9(10), 244; https://doi.org/10.3390/sym9100244
Received: 30 September 2017 / Revised: 11 October 2017 / Accepted: 15 October 2017 / Published: 21 October 2017
(This article belongs to the Special Issue Emerging Approaches and Advances in Big Data)
Over the years, advanced IT technologies have facilitated the emergence of new ways of generating and gathering data rapidly, continuously, and largely and are associated with a new research and application branch, namely, data stream mining (DSM). Among those multiple scenarios of DSM, the Internet of Things (IoT) plays a significant role, with a typical meaning of a tough and challenging computational case of big data. In this paper, we describe a self-adaptive approach to the pre-processing step of data stream classification. The proposed algorithm allows different divisions with both variable numbers and lengths of sub-windows under a whole sliding window on an input stream, and clustering-based particle swarm optimization (CPSO) is adopted as the main metaheuristic search method to guarantee that its stream segmentations are effective and adaptive to itself. In order to create a more abundant search space, statistical feature extraction (SFX) is applied after variable partitions of the entire sliding window. We validate and test the effort of our algorithm with other temporal methods according to several IoT environmental sensor monitoring datasets. The experiments yield encouraging outcomes, supporting the reality that picking significant appropriate variant sub-window segmentations heuristically with an incorporated clustering technique merit would allow these to perform better than others. View Full-Text
Keywords: data stream pre-processing; self-adaptive segmentation; clustering-based particle swarm optimization (CPSO); Internet of Things (IoT) datasets data stream pre-processing; self-adaptive segmentation; clustering-based particle swarm optimization (CPSO); Internet of Things (IoT) datasets
Show Figures

Figure 1

MDPI and ACS Style

Lan, K.; Fong, S.; Song, W.; Vasilakos, A.V.; Millham, R.C. Self-Adaptive Pre-Processing Methodology for Big Data Stream Mining in Internet of Things Environmental Sensor Monitoring. Symmetry 2017, 9, 244.

Show more citation formats Show less citations formats
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop