Abstract | ||
---|---|---|
We present a system for efficient detection, continuous maintenance and visualization of range-constrained optimal density clusters of moving objects trajectories, a.k.a. Continuous Maximizing Range Sum (Co-MaxRS) queries. Co-MaxRS is useful in any domain involving continuous detection of "most interesting" regions involving mobile entities (e.g., traffic monitoring, environmental tracking, etc.). Traditional MaxRS finds a location of a given rectangle R which maximizes the sum of the weighted-points (objects) in its interior. Since moving objects continuously change their locations, the MaxRS at a particular time instant need not be a solution at another time instant. Our system solves two important problems: (1) Efficiently computing Co-MaxRS answer-set; and (2) Visualizing the results. This demo will present the implementation of our efficient pruning schemes and compact data structures, and illustrate the end-user tools for specifying the parameters and selecting datasets for Co-MaxRS, along with visualization of the optimal locations. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1007/978-3-319-64367-0_29 | ADVANCES IN SPATIAL AND TEMPORAL DATABASES, SSTD 2017 |
Field | DocType | Volume |
Data structure,Data mining,Cluster (physics),Instant,Correlation clustering,Computer science,Visualization,Rectangle,Cluster analysis | Conference | 10411 |
ISSN | Citations | PageRank |
0302-9743 | 0 | 0.34 |
References | Authors | |
9 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Muhammed Mas-ud Hussain | 1 | 12 | 3.43 |
Goce Trajcevski | 2 | 1732 | 141.26 |
Kazi Ashik Islam | 3 | 2 | 0.72 |
Mohammed Eunus Ali | 4 | 260 | 31.28 |