This paper is aimed to propose a grid density clustering algorithm for spatio-temporal data that is based on the adaptation of the grid density based clustering algorithm. The algorithm is based on AGRID+ algorithm with 7 steps: partitioning, computing distance threshold, calculating densities, compensating densities, calculating density threshold (DT), clustering and removing noises. The adaptation is for the partitioning and calculating the distance threshold (r). The data utilized in this study is spatio-temporal fishery data located around the India Ocean from year 2000 until 2004. We utilized the fishery data in three types of aggregate , daily data, weekly data and monthy data. The result of this study shows that the time complexity for ST-AGRID is outperform the AGRID+. ST-AGRID improves the time complexity and at the same time maintains the accuracy. By utilizing the thresholding technique, clustering result of the ST-AGRID algorithm is identified as the potential fishing zone.
|Number of pages||14|
|Journal||International Journal of Software Engineering and its Applications|
|Publication status||Published - 2015|
- Grid-density based clustering
- Potential fishing zone
- Spatio-temporal clustering
- Temporal aggregate