Tis paper proposes a novel and efcient clustering algorithm for probability density functions based on k- -Medoids. Further, a scheme used for selecting the powerful initial medoids is suggested, which speeds up the computational time signifcantly. Also, a general proof for convergence of the proposed algorithm is presented. Te efectiveness and feasibility of the proposed algorithm are verifed and compared with various existing algorithms through both artifcial and real datasets in terms of adjusted Rand index, computational time, and iteration number. Te numerical results reveal an outstanding performance of the proposed algorithm as well as its potential applications in real life.