Ensembled Adaptive Fuzzy K-Means With Stochastic Extreme Gradient Boost Big Data Clustering on Geo-Social Networks

Anoop, M; Sripriya, P.

doi:10.1109/icacite51222.2021.9404574

Cited by 2 publications

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Clustering has known to be successful in obtaining or obtaining important information more quickly. Because of its ability and effectiveness in clustering data, the K-means technique is indisputably popular among the numerous clustering algorithms that already exist [3], [4]. K-means is a nonhierarchical data clustering method that aims to partition existing data into one or much more clusters [5], boosting clustering centroids to obtain optimal clustering centers [6], [7].…”

Section: Introductionmentioning

confidence: 99%

K-Means clustering of optimized wireless network sensor using genetic algorithm

Kadim¹,

Al-Mukhtar²,

Hasan³

et al. 2022

PEN

View full text Add to dashboard Cite

Wireless sensor network is one of the main technology trends that used in several different applications for collecting, processing, and distributing a vast range of data. It becomes an essential core technology for many applications related to sense surrounding environment. In this paper, a two-dimensional WSN scheme was utilized for obtaining various WSN models that intended to be optimized by genetic algorithm for achieving optimized WSN models. Such optimized WSN models might contain two cluster heads that are close to each other, in which the distance between them included in the sensing range, and this demonstrates the presence of a redundant number of cluster heads. This problem exceeded by reapplying the clustering of all sensors found in the WSN model. The distance measure was used to detect handled problem, while K-means clustering was used to redistributing sensors around the alternative cluster head. The result was extremely encouraging in rearranging the dispersion of sensors in the detecting region with a conservative method of modest number of cluster heads that acknowledge the association for all sensors nearby.

show abstract