Perbandingan Algoritme K-Means Dengan Algoritme Fuzzy C Means (FCM) Dalam Clustering Moda Transportasi Berbasis GPS

Rahman Syarif (Fakultas Ilmu Komputer, Universitas Brawijaya)
Muhammad Tanzil Furqon (Fakultas Ilmu Komputer, Universitas Brawijaya)
Sigit Adinugroho (Fakultas Ilmu Komputer, Universitas Brawijaya)

Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Vol 2 No 10 (2018): Oktober 2018

Transportation has become a basic necessity for today's society. But often the need for transportation is not followed by information on the availability of transportation in a certain place. In this case, data from GPS can be used to group the available modes of transportation and provide information on the number of each mode of transportation scattered in a certain place and time. Algorithm used to group modes of transportation in this research is K-Means and Fuzzy C Means (FCM). These two algorithms then compared which one with the better result. The transportation mode grouping on the K-Means algorithm is obtained from the smallest distance of the transport mode data with the cluster center. Whereas in the FCM algorithm, grouping is obtained from the greatest degree values. After 10 times testing, obtained an average of K-Means accuracy of 58.46154 and 70.86538 for FCM algorithm. While for the silhouette Coefficient value obtained an average of 0.4582670 for K-Means and 0.440682 for FCM algorithm. From the results, it was concluded that the FCM algorithm is superior to K-Means.