We propose a modified version of the Electrical Fuzzy C-Means (MEFCM) algorithm, enhanced with Coulomb’s Law to improve clustering performance. The algorithm is applied to seven datasets: Soybean (small), Dermatology, Breast Cancer, Iris, Wine, Ecoli, and Pima. Clustering quality is evaluated using Partition Coefficient (PC) and Partition Entropy (PE). Our results show that MEFCM performs particularly well on datasets with two clusters, delivering reliable and accurate results. For datasets with a higher number of clusters, convergence can be more challenging, and we suggest possible modifications to optimize the algorithm’s performance in such cases, providing insights for future improvements.