Parallelization of Partitioning Around Medoids (PAM) in K-Medoids Clustering on GPU

Abstract

K-medoids clustering is categorized as partitional clustering. K-medoids offers better result when dealing with outliers and arbitrary distance metric also in the situation when the mean or median does not exist within data. However, k-medoids suffers a high computational complexity. Partitioning Around Medoids (PAM) has been developed to improve k-medoids clustering, consists of build and swap steps and uses the entire dataset to find the best potential medoids. Thus, PAM produces better medoids than other algorithms. This research proposes the parallelization of PAM in k-medoids clustering on GPU to reduce computational time at the swap step of PAM. The parallelization scheme utilizes shared memory, reduction algorithm, and optimization of the thread block configuration to maximize the occupancy. Based on the experiment result, the proposed parallelized PAM k-medoids is faster than CPU and Matlab implementation and efficient for large dataset.

How heterogeneous is the dengue transmission profile in Brazil? A study in six Brazilian states

“…The clustering algorithm of choice was PAM (partitioning around medoids) because of its robustness, low sensitivity to outliers and its ability to preselect the number of clusters [ 40 , 41 ]. The elbow method was applied to give insight into the optimal number of cluster groups [ 42 ], while the silhouette method was used to assess the quality of the resulting partitions produced by PAM [ 41 – 43 ]. To guide the interpretation of the cluster analysis, summary statistics of the epi-features stratified by cluster were calculated, and maps of the classification were produced.…”

“…Epifeatures were standardized, with the aim of normalizing the data to establish a common scale, so that there is no distortion of values when there are larger ranges. The clustering algorithm of choice was PAM (partitioning around medoids) because of its robustness, low sensitivity to outliers and its ability to preselect the number of clusters [40,41]. The elbow method was applied to give insight into the optimal number of cluster groups [42], while the silhouette method was used to assess the quality of the resulting partitions produced by PAM [41][42][43].…”

Pattern Based Image Retrieval System by Using Clustering Techniques

“…It also has a high cluster accuracy compared to other distance-based partitioning algorithms for mixed variable data [22]. The PAM algorithm is used to reduce computational time at the swap step [25]. And also PAM is parallelization of on GPU.…”