Distance Function Analysis on Fuzzy C-Means for Clustering Aquaculture Production in North Sumatra

Abstrac As a maritime country, Indonesia has considerable fisheries resource potential, including in North Sumatra. However, less than optimal management causes a gap between the demand and supply of cultivated fish in this region. This research aims to analyze the use of the Fuzzy C-Means (FCM) method in classifying aquaculture production in North Sumatra based on Regency/City and primary commodities. This research also compares three methods of measuring Euclidean, Manhattan, and Chebyshev distances to determine the best method for data clustering. The North Sumatra Central Statistics Agency (BPS) data for 2018-2022. The results showed that there were two clusters, namely cluster 1 consisting of 4 districts with high fisheries production and cluster 2 consisting of 29 districts with low fisheries production. Evaluation using the Modified Partition Coefficient shows that the Manhattan distance method has the highest MPC value of 0.922. These results indicate that of the three distance methods, the Manhattan method is the best for applying Fuzzy C-Means to classify aquaculture production in North Sumatra.