Does Determination of Initial Cluster Centroids Improve the Performance of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="M1"><mml:mi>K</mml:mi></mml:math>-Means Clustering Algorithm? Comparison of Three Hybrid Methods by Genetic Algorithm, Minimum Spanning Tree, and Hierarchical Clustering in an Applied Study

Pourahmad, Saeedeh; Basirat, Atefeh; Rahimi, A. M.; Doostfatemeh, Marziyeh

doi:10.1155/2020/7636857

Cited by 10 publications

(6 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Pourahmad et al. (2020) compared the genetic algorithm, hierarchical clustering and minimum spanning tree hybrid initialization methods for KM clustering using internal, external and relative clustering performance measurements.…”

Section: Related Work Of Initialization Methodsmentioning

confidence: 99%

NDPD: an improved initial centroid method of partitional clustering for big data mining

Pandey

Shukla

2022

JAMR

View full text Add to dashboard Cite

PurposeThe K-means (KM) clustering algorithm is extremely responsive to the selection of initial centroids since the initial centroid of clusters determines computational effectiveness, efficiency and local optima issues. Numerous initialization strategies are to overcome these problems through the random and deterministic selection of initial centroids. The random initialization strategy suffers from local optimization issues with the worst clustering performance, while the deterministic initialization strategy achieves high computational cost. Big data clustering aims to reduce computation costs and improve cluster efficiency. The objective of this study is to achieve a better initial centroid for big data clustering on business management data without using random and deterministic initialization that avoids local optima and improves clustering efficiency with effectiveness in terms of cluster quality, computation cost, data comparisons and iterations on a single machine.Design/methodology/approachThis study presents the Normal Distribution Probability Density (NDPD) algorithm for big data clustering on a single machine to solve business management-related clustering issues. The NDPDKM algorithm resolves the KM clustering problem by probability density of each data point. The NDPDKM algorithm first identifies the most probable density data points by using the mean and standard deviation of the datasets through normal probability density. Thereafter, the NDPDKM determines K initial centroid by using sorting and linear systematic sampling heuristics.FindingsThe performance of the proposed algorithm is compared with KM, KM++, Var-Part, Murat-KM, Mean-KM and Sort-KM algorithms through Davies Bouldin score, Silhouette coefficient, SD Validity, S_Dbw Validity, Number of Iterations and CPU time validation indices on eight real business datasets. The experimental evaluation demonstrates that the NDPDKM algorithm reduces iterations, local optima, computing costs, and improves cluster performance, effectiveness, efficiency with stable convergence as compared to other algorithms. The NDPDKM algorithm minimizes the average computing time up to 34.83%, 90.28%, 71.83%, 92.67%, 69.53% and 76.03%, and reduces the average iterations up to 40.32%, 44.06%, 32.02%, 62.78%, 19.07% and 36.74% with reference to KM, KM++, Var-Part, Murat-KM, Mean-KM and Sort-KM algorithms.Originality/valueThe KM algorithm is the most widely used partitional clustering approach in data mining techniques that extract hidden knowledge, patterns and trends for decision-making strategies in business data. Business analytics is one of the applications of big data clustering where KM clustering is useful for the various subcategories of business analytics such as customer segmentation analysis, employee salary and performance analysis, document searching, delivery optimization, discount and offer analysis, chaplain management, manufacturing analysis, productivity analysis, specialized employee and investor searching and other decision-making strategies in business.

show abstract

Section: Related Work Of Initialization Methodsmentioning

confidence: 99%

NDPD: an improved initial centroid method of partitional clustering for big data mining

Pandey

Shukla

2022

JAMR

View full text Add to dashboard Cite

show abstract

“…To cluster data into coherent groups, the user must specify the optimal number of clusters ( k ) [15]. While data‐driven methods can help identify the optimal number of clusters [6], cluster selection should hinge predominantly on expert knowledge and interpretability of the results.…”

Section: Unsupervised Machine Learning Algorithmsmentioning

confidence: 99%

Unsupervised machine learning methods and emerging applications in healthcare

Eckhardt

Madjarova

Williams

et al. 2022

Knee Surg Sports Traumatol Arthrosc

View full text Add to dashboard Cite

Unsupervised machine learning methods are important analytical tools that can facilitate the analysis and interpretation of high-dimensional data. Unsupervised machine learning methods identify latent patterns and hidden structures in highdimensional data and can help simplify complex datasets. This article provides an overview of key unsupervised machine learning techniques including K-means clustering, hierarchical clustering, principal component analysis, and factor analysis. With a deeper understanding of these analytical tools, unsupervised machine learning methods can be incorporated into health sciences research to identify novel risk factors, improve prevention strategies, and facilitate delivery of personalized therapies and targeted patient care.

show abstract

“…and compared. The extensive review and comparative analysis 40,[64][65][66][67][68] provide decent references on KM initialization algorithms.…”

Section: Related Workmentioning

confidence: 99%

“…In addition to the aforementioned initialization algorithms, various initialization algorithms for KM clustering algorithms have been discovered and compared. The extensive review and comparative analysis 40,64–68 provide decent references on KM initialization algorithms.…”

Section: Related Workmentioning

confidence: 99%

“…Pourahmad et al 68 used the genetic algorithm, hierarchical clustering and minimum spanning tree hybrid methods to determine the initial centroid of the KM clustering and compared them to the standard KM initialization method through internal, external and relative clustering performance measurements. This comparative study discovered that standard KM clustering outperforms hybrid methods in terms of clustering performance, whereas hybrid methods only enhance convergence speed by reducing iteration.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Min‐max kurtosis stratum mean: An improved K‐means cluster initialization approach for microarray gene clustering on multidimensional big data

Pandey

Shukla

2022

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARY Microarray gene clustering is a big data application that employs the K‐means (KM) clustering algorithm to identify hidden patterns, evolutionary relationships, unknown functions and gene trends for disease diagnosis, tissue detection and biological analysis. The selection of initial centroids is a major issue in the KM algorithm because it influences the effectiveness, efficiency and local optima of the cluster. The existing initial centroid initialization algorithm is computationally expensive and degrades cluster quality due to the large dimensionality and interconnectedness of microarray gene data. To deal with this issue, this study proposed the min‐max kurtosis stratum mean (MKSM) algorithm for big data clustering in a single machine environment. The MKSM algorithm uses kurtosis for dimension selection, mean distance for gene relationship identification, and stratification for heterogeneous centroid extraction. The results of the presented algorithm are compared to the state‐of‐the‐art initialization strategy on twelve microarray gene datasets utilizing internal, external and statistical assessment criteria. The experimental results demonstrate that the MKSMKM algorithm reduces iterations, distance computation, data comparison and local optima, and improves cluster performance, effectiveness and efficiency with stable convergence.

show abstract

Does Determination of Initial Cluster Centroids Improve the Performance of K-Means Clustering Algorithm? Comparison of Three Hybrid Methods by Genetic Algorithm, Minimum Spanning Tree, and Hierarchical Clustering in an Applied Study

Cited by 10 publications

References 22 publications

NDPD: an improved initial centroid method of partitional clustering for big data mining

NDPD: an improved initial centroid method of partitional clustering for big data mining

Unsupervised machine learning methods and emerging applications in healthcare

Min‐max kurtosis stratum mean: An improved K‐means cluster initialization approach for microarray gene clustering on multidimensional big data

Contact Info

Product

Resources

About