Fuzzy C means application for area mapping of poor populations in Maluku Province: Efforts to reduce the number of poor populations

Based on data from Badan Pusat Statistik (BPS) in March 2016 stated that the number of poor people in Indonesia reached 28.01 million people. This figure is around 10.86 percent of the national population. The province of Maluku as the third poor contributor from all provinces in Indonesia reached 27.74 percent. Note that, there are 8 districts/cities in the Province of Maluku which are determined as disadvantaged areas, Maluku Barat Daya (MBD) is one of them. Based on data from BPS, in 2014 the percentage of poor people in the district of MBD was the second highest after the district of Maluku Tenggara Barat (MTB) of 11 districts/cities in Maluku where it reached 28.33 percent. The poverty level in a district is relatively difficult to reduce this is due to a large number of villages that have economic access isolation constraints due to unsupported geographical locations. Various programs and policies, both in the social, health, and so on, have been attempted to solve the problem, but they cannot overcome the existing poverty problem yet. The use of the Fuzzy C-Means (FCM) method to classify 11 districts/cities in Maluku Province based on data that is characteristic of poverty and consists of 10 variables. Grouping Process Using Number of Clusters: 5, Maximum iteration: 100, Expected velocity error: 10−5. The results using this method: grouping district/cities in Maluku based on poverty factors are: cluster 1 consists of Maluku Tenggara Barat, Buru, Kep. Aru, and Tual; cluster 2 consists of Maluku Barat Daya and Buru Selatan; cluster 3 consists of Maluku Tengah; cluster 4 consists of Maluku Utara, Seram Bagian Barat, and Seram Bagian Timur; and cluster 5 consists of Ambon. Each cluster describes the similarity level poverty based on Partition matrix U which loads the fuzzy centroid.