K-Affinity Propagation (K-AP) and K-Means Clustering for Classification of Earthquakes in Indonesia

Indonesia is located among three plate tectonics and ring of fire cause to be disaster prone especially to the earthquakes. Indonesia is the state of being having a level by earthquake high in the world as many as fifty nine thousand and eighty nine the earthquakes was happening in 2017.Hence, its efforts to mitigate is needed to reduce the impact caused by the earthquakes. Clustering K-Affinity Propagation and K-Means method used to cluster areas the earthquake, clustering aims to its efforts to mitigate. Testing validation use C-Index, Davies Bouldin Index, and Connectivity Index whereby the test was constructed using three and five clusters for K-Means and was constructed using two and four clusters using K-AP. Based on clusters of variance got that clusters of 4 is the sum clusters best has value ratio of the smallest. The results obtained the number of clusters one has two members with exemplar Celebes Sea, cluster two has twelve members with exemplar Halmahera, cluster three has one member with exemplar Minahassa Peninsulaand cluster four has thirty-four members with exemplar Sumba Region.