Transferring Semantic Categories with Vertex Kernels: Recommendations with Semantic SVD++ thumbnail
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Transferring Semantic Categories with Vertex Kernels: Recommendations with Semantic SVD++

Published on Dec 19, 20142051 Views

Matrix Factorisation is a recommendation approach that tries to understand what factors interest a user, based on his past ratings for items (products, movies, songs), and then use this factor inform

Related categories

Chapter list

TRANSFERRING SEMANTIC CATEGORIES WITH VERTEX KERNELS: RECOMMENDATIONS WITH SemanticSVD++00:00
Predicting Ratings 00:15
Latent Factor Models: Factor Consistency Problem01:31
Solution: Semantic Categories 02:35
Untitled03:29
Cold-start Categories03:56
Outline04:34
Dataset: MovieTweetings + URI Alignment04:59
Quantification: Cold-Start Categories Problem 06:55
Transferring Semantic Categories: Vertex Kernels 07:30
Category Transfer Function 08:13
User Profiling with Semantic Categories 09:31
Taste Evolution over Rated Categories 0.275 0.280 0.285 0.29010:15
Putting it all together: SemanticSVD++10:55
Transferring Categories using Vertex Kernels11:29
Experiments12:53
Experimental Setup 13:13
Results: Ratings Prediction Error 13:43
Conclusions14:33
Questions?15:09