Performance of the K-Means Algorithm for Water Quality Clustering

Siska Simamora; Paska Marto Hasugian

Authors

Siska Simamora Program Studi Teknologi Informasi, Universitas Putra Abadi Langkat, Sumatera Utara
Paska Marto Hasugian Program Studi Sains data, Fakultas Ilmu Komputer, Universitas Katolik Santo Thomas, Sumatera Utara

Keywords:

K-Means, Clustering, Distance Formula, Water Quality, Min-Max Normalization.

Abstract

Clustering is an unsupervised learning technique used to group data based on the degree of similarity among object characteristics. This study aims to analyze the application of a distance formula in cluster formation using the K-Means algorithm on the Water Quality Dataset. The dataset consists of 7,999 observations and 20 columns representing various water quality characteristics. The target column, is_safe, was removed, resulting in 19 features used in the clustering process. The preprocessing stages included checking for duplicate data, handling missing values, converting data into numerical format, and applying Min-Max normalization within the range of [0,1]. Normalization was performed to standardize the scale across features, ensuring that each feature contributed proportionally to the distance calculation. The clustering process was conducted using the K-Means algorithm, with data proximity determined based on the distance formula. The results indicate that data preprocessing and the selection of an appropriate distance formula are important factors in determining proximity patterns among objects and the resulting cluster formation. The use of normalized data can reduce the dominance of features with larger value ranges, thereby enabling the clustering process to represent data characteristics more proportionally. This study demonstrates that distance formula analysis plays an important role in supporting the formation of representative clusters in water quality data.

Downloads

Download data is not yet available.

References

World Health Organization, Guidelines for Drinking-Water Quality: Fourth Edition Incorporating the First and Second Addenda. Geneva, Switzerland: World Health Organization, 2022.

World Health Organization and United Nations Human Settlements Programme, Progress on the Proportion of Domestic and Industrial Wastewater Flows Safely Treated: Mid-term Status of SDG Indicator 6.3.1 and Acceleration Needs. Geneva, Switzerland: World Health Organization, 2024.

Food and Agriculture Organization of the United Nations, “Water quality,” FAO Land and Water, Rome, Italy.

F. Ghobadi and D. Kang, “Application of machine learning in water resources management: A systematic literature review,” Water, vol. 15, no. 4, Art. no. 620, 2023, doi: 10.3390/w15040620.

A. K. Jain, “Data clustering: 50 years beyond K-means,” Pattern Recognition Letters, vol. 31, no. 8, pp. 651–666, 2010.

R. Xu and D. Wunsch II, “Survey of clustering algorithms,” IEEE Transactions on Neural Networks, vol. 16, no. 3, pp. 645–678, May 2005.

J. MacQueen, “Some methods for classification and analysis of multivariate observations,” in Proc. 5th Berkeley Symp. Mathematical Statistics and Probability, vol. 1, 1967, pp. 281–297.

S. Lloyd, “Least squares quantization in PCM,” IEEE Transactions on Information Theory, vol. 28, no. 2, pp. 129–137, Mar. 1982.

J. Blömer, C. Lammersen, M. Schmidt, and C. Sohler, “Theoretical analysis of the k-means algorithm—A survey,” in Algorithm Engineering: Selected Results and Surveys, Lecture Notes in Computer Science, vol. 9220. Cham, Switzerland: Springer, 2016, pp. 81–116.

C. Wongoutong, “The impact of neglecting feature scaling in k-means clustering,” 2024. Penelitian ini membandingkan lima metode penskalaan fitur dan menunjukkan pentingnya feature scaling ketika K-Means diterapkan pada fitur dengan satuan yang berbeda.

P. J. Rousseeuw, “Silhouettes: A graphical aid to the interpretation and validation of cluster analysis,” Journal of Computational and Applied Mathematics, vol. 20, pp. 53–65, 1987.

D. L. Davies and D. W. Bouldin, “A cluster separation measure,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. PAMI-1, no. 2, pp. 224–227, Apr. 1979.

T. Zubaidah, N. Karnaningroem, and A. Slamet, “K-means method for clustering water quality status on the rivers of Banjarmasin, Indonesia,” ARPN Journal of Engineering and Applied Sciences, vol. 13, no. 11, pp. 3692–3697, 2018.

M. A. Novianta, S. Syafrudin, and B. Warsito, “K-Means clustering for grouping rivers in DIY based on water quality parameters,” JUITA: Jurnal Informatika, vol. 11, no. 1, pp. 155–163, 2023.

G. Arslan, C. Parim, and E. Cene, “Comparison of K-means and Fuzzy C-means clustering algorithms on water quality parameters: Case study of Ergene Basin for 17 stations,” in Current Debates on Natural and Engineering Sciences 9, pp. 145–159, 2023.