video thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Knowledge Discovery of Multiple-topic Document using Parametric Mixture Model with Dirichlet Prior

Published on 2007-09-143657 Views

Documents, such as those seen onWikipedia and Folksonomy, have tended to be assigned with multiple topics as a meta-data. Therefore, it is more and more important to analyze a relationship between a d

Presentation

Knowledge Discovery of Multiple-topic Document using Parametric Mixture Model with Dirichlet Prior 00:00
Contents 100:18
What kind of relationship between a document and topics exists?00:26
Contents 201:38
Probabilistic Generative Model01:42
PMM-part0103:14
PMM-part0204:14
Contents 304:45
Dirichlet distribution on mixture ratio (π1,π2,π3) 04:48
Estimate of Mixture Ratio06:09
Graphical Model09:03
Contents 409:42
Multiple-topic classification09:45
Evaluation by F-measure10:11
F-measure:PDMM vs PMM 1/211:39
Precision:PDMM vs PMM12:35
Recall:PDMM vs PMM12:47
F-measure:PDMM vs PMM 2/212:54
Contents 513:48
Word Ranking13:54
[Female(0.499)], [Male(0.460)] [Biological Markers(0.041)] 14:57
[Rats(0.411)], [Child(0.352)] [Incidence(0.237)]16:08
[Female(0.442)], [Animals(0.437)] [Pregnancy(0.066)],[Glucose(0.055)]16:23
[Pregnancy(0.502)],[Glucose(0.498)]17:11
Summary17:59
[Thank] [you] [for] [listening] [!]18:27