Sign inSign up
Leveraging User Libraries to Bootstrap Collaborative Filtering thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Leveraging User Libraries to Bootstrap Collaborative Filtering

Published on Oct 07, 20142665 Views

Laurent Charlin

We introduce a novel graphical model, the collaborative score topic model (CSTM), for personal recommendations of textual documents. CSTM's chief novelty lies in its learned model of individual librar

Research Sessions

Related categories

Data MiningKnowledge Extraction

Chapter list

Leveraging User Libraries to Bootstrap Collaborative Filtering00:00
Motivation00:06
Solution: Recommendations01:02
Data01:50
Desiderata02:25
Preference Prediction02:59
Preference Prediction with Side Information04:05
Collaborative Score Topic Model CSTM - 104:49
Collaborative Score Topic Model CSTM - 206:28
Collaborative Score Topic Model CSTM - 307:47
Collaborative Score Topic Model CSTM - 409:00
Collaborative Score Topic Model CSTM - 509:43
CSTM10:17
Related Work10:54
Experiments11:28
Datasets - 111:32
Datasets - 212:20
Preference prediction results (ICML'12)13:16
Book recommendation results13:24
Preference Prediction with Textual Side Information13:26
NIPS'13 recommendation system14:09
Conclusion & Future Work14:16