—The aim of this paper is to analyze the determination of the potential fishing zones based on data mining approach. The algorithm utilized in this study is AGRID+, a grid density based clustering for high dimensional data. The case study area is in eastern Indian Ocean located at 16.56 - 2 S and 100.49 – 140 E. The algorithm is implemented in 7 phases, partitioning, computing distance threshold, calculating densities, compensating densities, calculating density threshold, clustering and removing noise. The clustering result is evaluated by the Silhouette index. The results of the study show that the best cluster formed at daily aggregate temporal with number of cell (m
) = 14 and the number of cluster formed was 50 clusters. The constant execution time is in line with the increasing the value of m
. From three different temporal aggregate, the daily aggregate is running relatively constant for various m
value. To determine the potential fishing zones for different temporal aggregate can be achieved by applying the thresholding technique to the cluster result. Utilizing the data mining approach yielded a prominent 22 daily clusters identified as potential fishing zone.
—AGRID+, data mining, clustering, potential
fishing zone, spatio-temporal.
The authors are with the Faculty of Computer Science, Universitas
Indonesia, Depok 16424 Indonesia. Devi Fitrianah is also with the Faculty of
Computer Science, Universitas Mercu Buana, Jakarta 11650 Indonesia
(e-mail: firstname.lastname@example.org, email@example.com, firstname.lastname@example.org,
Cite: Devi Fitrianah, Hisyam Fahmi, Achmad Nizar Hidayanto, and Aniati Murni Arymurthy, "A Data Mining Based Approach for Determining the Potential Fishing Zones," International Journal of Information and Education Technology vol. 6, no. 3, pp. 187-191, 2016.