Enhancing Clustering with Automated Quality Assessment of Spanning Tree Method in Real Data Analysis

This paper presents the development of an automated system for assessing the quality of the spanning tree method in clustering real data tuples. The system is designed to enhance the efficiency and accuracy of data clustering, allowing researchers and data analysts to uncover meaningful patterns within large datasets. The software reads data from Excel files, constructs a minimal spanning tree using Prim's algorithm, clusters the tree, and evaluates clustering quality using the Davies-Bouldin Index and the SV Index. The program's userfriendly interface simplifies the processes of loading data, constructing index plots, and saving clustering results. Experiments conducted on different datasets demonstrate the program's accuracy in reading data, saving clustering variants, and calculating quality indices. The implementation of parallelized algorithms for constructing the minimal spanning tree significantly improves performance, enabling the processing of large datasets efficiently. This automated system provides a robust tool for the quality assessment of the spanning tree method in clustering real data tuples, with broad applicability across various domains such as medicine, biology, physics, and economics. Future enhancements may include additional clustering methods, improved visualization features, and a web-based version of the program to facilitate global collaboration among researchers and analysts.