Modeling Human Location Data with Mixtures of Kernel Densities thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Modeling Human Location Data with Mixtures of Kernel Densities

Published on Oct 07, 20141905 Views

Location-based data is increasingly prevalent with the rapid increase and adoption of mobile devices. In this paper we address the problem of learning spatial density models, focusing specifically on

Related categories

Chapter list

Modeling Human Location Data with Mixtures of Kernel Densities00:00
Treat locations as events - 100:18
Treat locations as events - 200:34
Treat locations as events - 300:45
Treat locations as events - 400:48
Motivation00:54
Kernel Density Estimation - 101:50
Kernel Density Estimation - 202:05
Kernel Density Estimation - 302:12
Kernel Density Estimation - 402:35
Kernel Density Estimation - 503:03
Kernel Density Estimation - 603:33
Kernel Density Estimation - 703:47
A tale of Two Cities - 103:48
A tale of Two Cities - 204:11
Adaptive Bandwidth KDE - 105:28
Adaptive Bandwidth KDE - 205:57
Adaptive Bandwidth KDE - 306:02
Adaptive Bandwidth KDE - 406:14
Comparing Methods - 106:29
Comparing Methods - 206:58
Comparing Methods - 307:22
Comparing Methods - 407:45
Comparing Methods - 509:08
Mixture of Kernel Densities - 109:40
Mixture of Kernel Densities - 210:35
Evaluation10:50
Evaluation – Log Likelihood - 111:01
Evaluation – Log Likelihood - 211:15
Evaluation – Log Likelihood - 311:33
Evaluation – Log Likelihood - 411:42
Evaluation – Log Likelihood - 511:53
Evaluation – Log Likelihood - 611:59
Evaluation – Fraud Detection - 112:07
Evaluation – Fraud Detection - 212:20
Evaluation – Fraud Detection - 312:40
Evaluation – Fraud Detection - 412:55
Evaluation – Fraud Detection - 513:32
Conclusions13:44
Future Work14:01
Acknowledgements14:25
Q & A14:36