Performance analysis in text clustering using k-means and k-medoids algorithms for Malay crime documents

<span lang="EN-US">Few studies on text clustering for the Malay language have been conducted due to some limitations that need be addressed. The purpose of this article is compare two algorithms k-means and k-medoids using Euclidean distance similarity determine which method best documents. Both are applied 1000 documents pertaining housebreaking crimes involving a variety different modus operandi. Comparability results indicate algorithm performed at relevant documents, with 78% accuracy rate. K-means also achieves performance cluster evaluation when comparing average within-cluster algorithm. However, perform exceptionally well Davis Bouldin index (DBI). Furthermore, dependent number initial clusters, where appropriate can determined elbow method.</span>