The existence of identical objects in a data set is a necessity. This paper proposes a new indicator and procedure to obtain the initial medoids. The new algorithm guarantees no empty groups and identical objects in the same group, either in the initial or final groups. We use six real data sets to evaluate the proposed method and compare the results of other methods in terms of adjusted Rand index and clustering accuracy. The experiment results show that the performance of the proposed method is comparable with other methods
                        
                        
                        
                        
                            
                                Copyrights © 2022