Extracting Hotspots without A-priori by Enabling Signal Processing over Geospatial Data
Hotspot (geology)
Location-based service
DOI:
10.1145/3139958.3140002
Publication Date:
2018-09-13T12:54:52Z
AUTHORS (4)
ABSTRACT
The proliferation of mobile devices equipped with internet connectivity and global positioning functionality (GPS) has resulted in the generation large volumes spatiotemporal data. This led to rapid evolution location-based services. anticipatory nature these services, demand exploitation a broader range user information for service personalization. Determining users' places interest, i.e. hotspots is critical understand their behaviors preferences. Existing techniques detect rely on set a-priori determined parameters that are either dataset dependent or derived without any empirical basis. leads biased results inaccuracies estimating total number belonging user, shape average dwelling time. In this paper, we propose parameter-less technique extracting from trajectories assumptions. We eliminate parameter dependence by treating as signals signal processing algorithms derive hotspots. experimentally show that, our does not necessitate behavior bounds, which makes it suitable extract larger variety datasets across users having disparate mobility behaviors. Our evaluation real world dataset, accuracy rates exceeding 80% outperforms traditional clustering used hotspot detection.
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES (15)
CITATIONS (5)
EXTERNAL LINKS
PlumX Metrics
RECOMMENDATIONS
FAIR ASSESSMENT
Coming soon ....
JUPYTER LAB
Coming soon ....