Analisis Algoritma Clustering Dalam Kasus Penentuan Jenis Bunga Iris

Diwahana Mutiara Candrasari (Sekolah Tinggi Ilmu Komputer Yos Sudarso Purwokerto)

Media Aplikom
Vol 9 No 2 (2017)

Clustering is one process of data mining that aims to partition existing data into one or more cluster objects based on the characteristics it has. Data with the same characteristics are grouped in one cluster and data with different characteristics are grouped into another cluster. In this study will perform comparation and analyze the best algorithm for categorize flowers by using iris dataset. Clustering algorithm techniques used include K-means, and K-medoids,. The value of davies bouldin and number of clusters will be investigated using the rapidminer tool. The results show that the K-Means algorithm has the lowest davies bouldin value of 0.167, while K-Medoids yields davies bouldin value of 0.291, but among the three algorithms, the K-Means algorithm is the most dominant and best in the comparative process of grouping iris flowers. Keywords: K-means, K-medoids , clustering