Resource Load Balancing on Cloud Infrastructure for Subscriber Management in Comparison with Raw Unbalanced Data for Calculation of Energy Consumption

V. Venkatesh, A. Shri Vindhya

2023

Abstract

This study compares the novel K-Means clustering method against the more popular Expectation-Maximization Clustering technique in order to envision whether one produces more accurate results when used to partition patrons' online social network activity. Materials and Methods: Extensive testing was conducted to determine the accuracy percentages of both the K-Means clustering method and the Expectation-Maximization clustering algorithm. The sample size used for each test was 110, and for the Expectation-Maximization algorithm, a G power (value) of 0.6 was employed. Results: According to the results, the novel K-Means clustering approach is superior to the Expectation-Maximization Clustering methodology in terms of accuracy (87.97% vs. 79.77%). At a significance level of 0.001 (p < 0.05), the data strongly indicates a noteworthy distinction between the two groups. When compared to the Expectation-Maximization clustering approach, the novel K-Means technique fared very well.

Paper Citation

in Harvard Style

Venkatesh V. and Shri Vindhya A. (2023). Resource Load Balancing on Cloud Infrastructure for Subscriber Management in Comparison with Raw Unbalanced Data for Calculation of Energy Consumption. In Proceedings of the 1st International Conference on Artificial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics - Volume 1: AI4IoT; ISBN 978-989-758-661-3, SciTePress, pages 354-359. DOI: 10.5220/0012772100003739

in Bibtex Style

@conference{ai4iot23,
author={V. Venkatesh and A. Shri Vindhya},
title={Resource Load Balancing on Cloud Infrastructure for Subscriber Management in Comparison with Raw Unbalanced Data for Calculation of Energy Consumption},
booktitle={Proceedings of the 1st International Conference on Artificial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics - Volume 1: AI4IoT},
year={2023},
pages={354-359},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012772100003739},
isbn={978-989-758-661-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Artificial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics - Volume 1: AI4IoT
TI - Resource Load Balancing on Cloud Infrastructure for Subscriber Management in Comparison with Raw Unbalanced Data for Calculation of Energy Consumption
SN - 978-989-758-661-3
AU - Venkatesh V.
AU - Shri Vindhya A.
PY - 2023
SP - 354
EP - 359
DO - 10.5220/0012772100003739
PB - SciTePress