Chaotic Association Feature Extraction of Big Data Clustering Based on the Internet of Things

Chaotic Association Feature Extraction of Big Data Clustering Based on the Internet of Things

Extraction

Big data

Energy utilization

Feature extraction

Cluster analysis

Internet of things

Data mining

Clustering algorithms

Phase space methods

This article addresses the stabilization of chaotic characteristics in abnormal data by proposing chaotic correlation feature extraction of big data clustering based on the Internet of things. The chaotic features in big data usually show complex folding and distortion without obvious rules and order and non-synchronization. In this article, the dimension of extracted correlation is utilized as the chaotic feature for the clustering of big data. The one-dimensional time series that can be extended in multi-dimensional space is analysed based on phase space reconstruction, to extract the chaotic correlation dimension (CCD) features. After the relevant experimental analysis, this paper mainly compares the energy consumption and processing time of the two respective algorithms. In the simulation parameter design, the time interval of big data packet generation is 0.1s, and the data is generated from the simulation time of 300s. The results obtained show that when dealing with the same amount of data, the energy consumption of this algorithm is significantly lower than that of the traditional algorithm. When dealing with the same amount of data, the time required by this algorithm is significantly lower than that of the traditional algorithm. This is because this algorithm is easy to implement and has good clustering efficiency for data, so the clustering time is short. With the gradual increase in the amount of data, the correlation dimension of this algorithm tends to be stable. While the correlation dimension of the traditional algorithm fluctuates greatly, it is revealed that the proposed approach has high data clustering efficiency and verifies the effectiveness of this algorithm. 2022 Slovene Society Informatika. All rights reserved.

333-342

3

46

Liu, Xiaoming

Singh, Thipendra Pal

Gupta, Rajeev Kumar

Onyema, Edeh Michael

Informatica (Slovenia)

2022

journalArticle

3505596

10.31449/inf.v46i3.3943

http://dx.doi.org/10.31449/inf.v46i3.3943