@article {10.3844/jcssp.2025.2074.2080,
article_type = {journal},
title = {Adaptive Data Transformation for Enhanced Clustering Performance in Diagnostic Systems},
author = {Al-Batah, Mohammed Subhi},
volume = {21},
number = {9},
year = {2025},
month = {Oct},
pages = {2074-2080},
doi = {10.3844/jcssp.2025.2074.2080},
url = {https://thescipub.com/abstract/jcssp.2025.2074.2080},
abstract = {This paper presents an enhanced approach to the Balanced Iterative Reducing and Clustering using Hierarchies (BIRCH) algorithm, aimed at improving clustering accuracy in medical data, specifically for breast cancer diagnosis. The proposed method introduces a modified data transformation technique to optimize the original BIRCH algorithm. This transformation refines the clustering process, resulting in significant improvements in diagnostic accuracy. The modified BIRCH algorithm was tested on a breast cancer dataset and achieved a clustering accuracy of 98.40%, a substantial improvement compared to 33.22% accuracy obtained using the original algorithm. Experimental results demonstrate that the use of transformed data not only enhances the performance of BIRCH but also highlights its effectiveness in scenarios with two clusters and a threshold value of two. These findings suggest that data transformation plays a critical role in refining hierarchical clustering algorithms, offering better diagnostic insights in medical applications.},
journal = {Journal of Computer Science},
publisher = {Science Publications}
}