PERBANDINGAN PENERAPAN ALGORITMA K-MEANS DAN FUZZY C-MEANS DALAM ANALISIS CLUSTERING TERHADAP PERGERAKAN HARGA HISTORIS SAHAM BANK RAKYAT INDONESIA

Abstract

This study aims to analyze and compare the application of K-Means and Fuzzy C-Means algorithms for clustering historical stock price movements of Bank Rakyat Indonesia (BRI). Clustering is a method that groups data based on similarities, crucial in stock data analysis to aid more precise investment decision-making. The K-Means algorithm deterministically assigns each data point to a single cluster, while Fuzzy C-Means allows partial membership across multiple clusters, offering greater flexibility. The research findings indicate that the K-Means algorithm forms three primary clusters with a Silhouette Score of 0.4667, which defines clusters more clearly than Fuzzy C-Means, which has a score of 0.4199. The clusters produced by K-Means provide better-defined separations among stocks with medium, high, and low prices, based on price movements and transaction volume. In contrast, Fuzzy C-Means, despite its ability to handle overlapping data, results in less clearly defined clusters compared to K-Means. Based on these results, the K-Means algorithm is deemed more effective for clustering analysis in the context of BRI stocks. This research is expected to contribute to the development of more comprehensive stock movement analysis models and support investors in making better-informed investment decisions.