en
0.25
0.5
0.75
1.25
1.5
1.75
2
Non-Parametric Scan Statistics for Event Detection and Forecasting in Heterogeneous Social Media Graphs
Published on 2014-10-072343 Views
Event detection in social media is an important but challenging problem. Most existing approaches are based on burst detection, topic modeling, or clustering techniques, which cannot naturally model t
Related categories
Presentation
Non-Parametric Scan Statistics for Event Detection and Forecasting in Heterogeneous Social Media Graphs00:00
Why Can We Detect & Forecast Events from Social Media?00:14
Disease Event Signals on Twitter01:03
Elephant And The Blind Men - 102:11
Elephant And The Blind Men - 202:35
Elephant And The Blind Men - 302:45
Twitter Heterogeneous Network03:27
Twitter Heterogeneous Network (Example)04:20
Node Attributes04:36
Research Questions04:54
Summary of Our Major Contributions05:31
Two Stage Empirical Calibration Process - 106:04
Two Stage Empirical Calibration Process - 207:08
Two Stage Empirical Calibration Process - 308:25
Nonparametric Scan Statistics08:53
Berk-Jones (BJ) Statistic10:17
Nonparametric Graph Scanning - 111:09
Nonparametric Graph Scanning - 211:21
Nonparametric Graph Scanning - 311:38
Nonparametric Graph Scanning - 411:42
Nonparametric Graph Scanning - 511:51
Nonparametric Graph Scanning - 612:16
Nonparametric Graph Scanning Algorithm - 112:18
Nonparametric Graph Scanning Algorithm - 212:28
Nonparametric Graph Scanning Algorithm - 312:29
Nonparametric Graph Scanning Algorithm - 412:31
Experimental Evaluations13:25
Experiment Settings13:42
Twitter Dataset for Hantavirus Outbreaks14:09
Twitter Dataset for Civil Unrests14:23
Comparison with Baseline Methods - 114:43
Comparison with Baseline Methods - 214:59
Conclusion15:20