Scalable Text and Link Analysis with Mixed-Topic Link Models thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Scalable Text and Link Analysis with Mixed-Topic Link Models

Published on Sep 27, 20133343 Views

Many data sets contain rich information about objects, as well as pairwise relations between them. For instance, in networks of websites, scientific papers, and other documents, each node has content

Related categories

Chapter list

Scalable Text and Link Analysis with Mixed-Topic Link Models00:00
General form of the problem00:16
Our work00:46
Motivation - 101:28
Motivation - 203:03
Motivation - 303:19
Motivation - 403:34
Probabilistic Latent Semantic Analysis (plsa) [?]04:25
bkn model [Ball, Karrer, Newman (2011)] - 106:01
bkn model [Ball, Karrer, Newman (2011)] - 206:54
bkn model [Ball, Karrer, Newman (2011)] - 307:34
Graphical model - 108:24
Graphical model - 208:59
Likelihood functions - 109:39
Likelihood functions - 210:06
Likelihood functions - 310:14
Expectation-Maximization(EM) algorithm10:37
Time complexity of the EM algorithm - 111:45
Time complexity of the EM algorithm - 212:28
Real-world data sets12:35
Convergence curves13:35
Running time on real-world data sets14:21
Discrete labels and local search15:28
Performance on document classification - 116:28
Performance on document classification - 217:21
Performance on document classification - 317:45
Performance on document classification - 417:56
Balancing content and links18:12
Performance on document classification18:49
Performance on link prediction20:19
Conclusions20:45