Jointly Modeling Aspects, Ratings and Sentiments for Movie Recommendation (JMARS) thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Jointly Modeling Aspects, Ratings and Sentiments for Movie Recommendation (JMARS)

Published on Oct 07, 20143082 Views

Recommendation and review sites offer a wealth of information beyond ratings. For instance, on IMDb users leave reviews, commenting on different aspects of a movie (e.g. actors, plot, visual effects),

Related categories

Chapter list

Jointly Modeling Aspects, Ratings and Sentiments for Movie Recommendation00:00
Movie review - 100:10
Movie review - 200:22
Movie review - 300:28
Movie review - 400:38
Aspect sentiment modeling - 101:24
Aspect sentiment modeling - 201:55
Everything but the kitchen sink - 102:11
Everything but the kitchen sink - 202:32
Aspect modeling02:47
Rating modeling03:51
Language model04:43
Topic model06:01
Sentiment Mapping06:30
Inference by hybrid sampling / optimization06:56
Experiments - 107:56
Experiments - 208:00
Mean squared error (MSE)08:25
Perplexity09:10
Cold start09:29
Qualitative evaluation10:08
Sentiment & movie-specific words10:11
Aspect rating10:27
Aspect & Aspect-sent words11:01
Movie specific Background tokens11:39
Summary12:06