Peningkatan Kinerja Algoritma K-Means Dengan Teknik Pillar untuk Penentuan Centroid

Abstract

The K-Means algorithm is a popular clustering technique used in many applications, including machine learning, data mining, and image processing. Despite its popularity, the algorithm has several limitations, including sensitivity to centroid value initialization and clustering quality. In this paper, the authors propose a new technique called the "pillar technique" to improve the performance of the K-Means algorithm. The pillar technique involves dividing a dataset into smaller sub-datasets, calculating the centroids for each sub-dataset, and then combining the centroids to get the final cluster centroid. The authors compared the performance of the K-Means algorithm with and without pillar techniques on several datasets. The author's results show that the Pillar technique improves the quality of grouping with a difference in Sum of Square Error (SSE) values by up to 50% with the acquisition of SSE K-Means of 50.07678 and K-Means with the Pillar technique of 25.09753. The author's findings show that the Pillar technique is an effective method to improve the performance of the K-Means algorithm, especially in research using Baitul Mal wa Tamwil (BMT) customer data in Batang Kuis.