Clustering Methods for Network Data Analysis in Programming

Abstract

In the modern world, data volumes are constantly increasing, and clustering has become an essential tool for identifying patterns and regularities in large datasets. The relevance of this study is associated with the growing need for effective data analysis methods in programming. The objective is to evaluate different clustering techniques within the programming domain and explore their suitability for analysing a wide range of datasets. Inductive and deductive methodologies, concrete illustrations, and visual techniques were employed. The clustering techniques were implemented using RStudio and Matlab tools. The study's findings facilitated the identification of crucial attributes of clustering techniques, including hierarchical structure, cluster quantity, and similarity metrics. The application of several data analysis and visualisation approaches, including k-means, c-means, hierarchical, least spanning tree, and linked component extraction, was illustrated. This study elucidated the clustering approaches that may be optimally employed in various contexts, resulting in enhanced precision in analyses and data-informed decision-making. The study's practical significance is in enhancing programmers' methodological toolset with tools for data analysis and processing.