Improved partitioning technique for density cube-based spatio-temporal clustering method

This work proposes a novel partitioning technique on the density-cube-based data model for Spatio-temporal clustering method. further adapts this approach to data. We have compared IMSTAGRID-the proposed algorithm ST-DBSCAN, AGRID+, and ST-AGRID algorithms found that IMSTAGRID improves interval expansion is able achieve uniformity in spatial temporal dimensional values. Three types of sets been used experiment: storm set two synthetic – 1 2. Both 2 were comparable terms scattering points, while contained clustered The performance method was measured via silhouette analysis, its results surpassed other investigated; index 0.970, 0.993 using 1. also outperformed baseline (ST-DBSCAN, ST-AGRID) labeling accuracy set, yielding 82.68%, 38.36%, 76.13%, 78.66%, respectively.