PERBANDINGAN ALGORITMA K-MEANS DAN K-MEDOIDS DALAM PENGELOMPOKKAN BANK UMUM SYARIAH DI INDONESIA DENGAN DAVIES BOULDIN INDEKS, CALINSKI HARABASZ INDEKS & KOEFISIEN SILHOUTTE

Abstract

Financial ratios are a tool that is often used in examining the financial performance of a bank. There is still a gap in the finances regarding Islamic commercial banks in Indonesia, as seen in one of the ratios that describes the financial performance of a bank. The Return On Assets ratio in 2022 will have a very different gap. This means that the performance of shariah commercial banks is still not good and optimal. In this research, sharia commercial banks in Indonesia will be grouped. The process of identifying the financial ratios of each Islamic commercial bank in Indonesia is carried out using cluster analysis, namely the K-Means and K-Medoids methods. The two methods used are different, so validation needs to be carried out to assess the grouping results using three validations, namely the Davies Bouldin Index, Calinski Harabasz Index, and Silhoutte Coefficient, to produce the best cluster structure from the resulting grouping can be known. Based on the 3 validation clusters used, namely DBI, CHI, and Silhoutte Coefficient, in a comparison of the two K-Means and K-Medoids algorithms, it was concluded that the K-Means algorithm had the best results in grouping Islamic commercial banks.