Minimum spanning tree‐based cluster analysis: A new algorithm for determining inconsistent edges

Abstract

In recent years, graph‐based data clustering algorithms have become popular as they perform connectivity‐based rather than centroid‐based partitioning. Methods related to minimum spanning tree (MST)‐based data clustering are types of graph‐based algorithms that can recognize arbitrary shapes of clusters by eliminating inconsistent edges from MST graphs. In all MST‐based data clustering algorithms, definition of inconsistent edges is the main problem that needs to be addressed. The longest edges in MST graphs are considered as inconsistent edges under ideal conditions. Nevertheless, outliers often exist in real‐world tasks, which makes the longest edges inaccurate cluster separation indicators. In this paper, we propose a new data clustering algorithm using MST and a critical distance method. The proposed algorithm solves the main issue of MST‐based data clustering, namely identifying and removing inconsistent edges to obtain clusters even in the event that the dataset contains some outliers. It begins by constructing the MST over a given weighted graph based on Euclidean distance and then splits up the graph into clusters by eliminating inconsistent edges using critical distance as a threshold. Integration of the advantages of both MST and critical distance methodology to obtain optimal clusters is the main contribution of this work. The conducted experimental analysis and results using different datasets prove that our proposed clustering algorithm yields better overall performance compared with the most common data clustering algorithms. Taking the Liver and Tumor datasets as an example, the proposed algorithm outperforms all other clustering algorithms with clustering accuracy equal to 0.579 and 0.660, respectively.

Development of Gluten-Free Bread Production Technology with Enhanced Nutritional Value in the Context of Kazakhstan

“…The results of organoleptic indicators of flours are shown in Table 5. The results meet the general standards [65][66][67] for all flours. From the data in Table 5, it can be seen that green buckwheat, corn, and plantain flour showed high scores for all organoleptic indicators.…”

“…In this method, objects are grouped into hierarchical clusters, wherein each cluster comprises the most similar objects, and higher-level clusters encapsulate those from lower levels. Following an agglomerative approach, the clustering begins with individual objects and progressively merges the closest pairs iteratively, forming increasingly larger clusters until all objects are within a unified cluster [65,66].…”

Neighboring-Aware Hierarchical Clustering

“…The target application has a significant impact on how well the clustering method and similarity metric perform together. Finally, it is important to note that the authors did not include MST-based techniques in general (Şaar & Topcu, 2022) or those that use metaheuristics or evolutionary algorithms (Halim & Uzma, 2018), as a comprehensive analysis of data clustering algorithms is outside the purview of this work. Therefore, the authors included and discussed the research that is most important to the study in this paper.…”

Special issue on High Performance Computing Conference (BASARIM‐2020)

“…Saar et al 8 present a new algorithm for solving the main problem faced in processes of minimum spanning tree (MST)-based data clustering:…”

“…Saar et al 8 present a new algorithm for solving the main problem faced in processes of minimum spanning tree (MST)‐based data clustering: the identification and removal of inconsistent edges to successfully obtain clusters even when datasets contain outliers. Their new algorithm for data clustering uses MST in combination with a critical distance method.…”