Efficiency Evaluation of K-Means, K-Medians, and K-Mode Clustering Methods Using SSD

Bounmy PHANTHAVONG , Soulith SENGMANOTHAM , Sommany LUSAVONG , phonesouda souphamith

K-Means, K-Medians, K-Mode, Sum of Squares Distance (SSD), Elbow method

This study compares performance of three data clustering algorithms: K-Means, K-Medians, and K-Mode. Using correlation analysis, key variables with the highest interrelationships were identified and then used to determine the optimal number of clusters through the Elbow method. Once the optimal cluster count was established, the clustering was conducted using the three methods, and the efficiency was evaluated based on the Sum of Squares Distance (SSD). A lower SSD indicates a more efficient clustering result. The analysis was performed on test data from 15,602 students across four subjects Mathematics, Physics, Lao Language, and Grography. Physics and Lao Language were found to be the most highly correlated variables. Clustering the data using K-Means, K-Medians, and K-Mode produced SSD values of 0.5535, 0.7476, and 1.4937, respectively. The results demonstrate the K-Means achieved the best performance, delivering the most efficient clustering with the lowest SSD. In conclusion, K-Means outperforms K-Medians and K-Mode, making it the most effective algorithm for clustering the given dataset

Published Paper   E-Certificate

"Efficiency Evaluation of K-Means, K-Medians, and K-Mode Clustering Methods Using SSD", IJSDR - International Journal of Scientific Development and Research (www.IJSDR.org), ISSN:2455-2631, Vol.9, Issue 11, page no.136 - 145, November-2024, Available :https://ijsdr.org/papers/IJSDR2411018.pdf

Volume 9 Issue 11, November-2024

Pages : 136 - 145

Paper Reg. ID: IJSDR_212669

Published Paper Id: IJSDR2411018

Downloads: 000346998

Research Area: Science & Technology

Country: Vientiane, Vientiane Capital, Lao People's Democratic Republic

Published Paper PDF: https://ijsdr.org/papers/IJSDR2411018

Published Paper URL: https://ijsdr.org/viewpaperforall?paper=IJSDR2411018

ISSN: 2455-2631 | IMPACT FACTOR: 9.15 Calculated By Google Scholar | ESTD YEAR: 2016

An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 9.15 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator

Publisher: IJSDR(IJ Publication) Janvi Wave

This work is licensed under a Creative Commons Attribution 4.0 International License and The Open Definition