filmov
tv
RecSys 2016: Paper Session 8 - Meta-Prod2Vec: Product Embeddings Using Side-Information
Показать описание
Flavian Vasile, Elena Smirnova, Alexis Conneau
We propose Meta-Prod2vec, a novel method to compute item similarities for recommendation that leverages existing item metadata. Such scenarios are frequently encountered in applications such as content recommendation, ad targeting and web search. Our method leverages past user interactions with items and their attributes to compute low-dimensional embeddings of items. Specifically, the item metadata is injected into the model as side information to regularize the item embeddings. We show that the new item representations lead to better performance on recommendation tasks on an open music dataset.
We propose Meta-Prod2vec, a novel method to compute item similarities for recommendation that leverages existing item metadata. Such scenarios are frequently encountered in applications such as content recommendation, ad targeting and web search. Our method leverages past user interactions with items and their attributes to compute low-dimensional embeddings of items. Specifically, the item metadata is injected into the model as side information to regularize the item embeddings. We show that the new item representations lead to better performance on recommendation tasks on an open music dataset.
0:18:03
RecSys 2016: Paper Session 8 - Convolutional Matrix Factorization for Context-Aware Recommendation
0:19:13
RecSys 2016: Paper Session 8 - Meta-Prod2Vec: Product Embeddings Using Side-Information
0:21:04
RecSys 2016: Paper Session 8 - Parallel Recurrent Neural Network Architectures for Recommendations
0:22:06
RecSys 2016: Paper Session 9 - TAPER: A Contextual Tensor-Based Approach
0:10:54
RecSys 2016: Paper Session 9 - Modelling Contextual Information in Session-Aware Recommender Systems
0:16:11
RecSys 2016: Paper Session 7 - Past, Present, & Future of Recommender Systems: Industry Perspect...
0:15:46
RecSys 2016: Paper Session 9 - The Contextual Turn: from Context-Aware to Context-Driven
0:10:24
RecSys 2016: Paper Session 6 - Optimizing Similar Item Recommendations
0:09:43
RecSys 2016: Paper Session 1 - Contrasting Offline and Online Results when Evaluating Recommendation
0:18:51
RecSys 2016: Paper Session 7 - Behaviorism is Not Enough: Better Recommendations
0:19:53
RecSys 2016: Paper Session 11 - Using Navigation to Improve Recommendations in Real-Time
0:15:57
RecSys 2016: Paper Session 7 - Algorithms Aside: Recommendation As The Lens Of Life
0:14:45
RecSys 2016: Paper Session 11 - Bayesian Personalized Ranking with Multi-Channel User Feedback
0:17:59
RecSys 2016: Paper Session 10 - Vista: A Visually, Socially, and Temporally-aware Model
0:18:29
RecSys 2016: Paper Session 10 - STAR: Semiring Trust Inference for Trust-Aware Social Recommenders
0:17:49
RecSys 2016: Paper Session 11 - Personalized Recommendations using Knowledge Graphs
0:14:05
RecSys 2016: Paper Session1 - Recommender Systems Self Actualization
0:09:39
RecSys 2016: Paper Session 1 - Adaptive, Personalized Diversity for Visual Discovery
0:20:13
RecSys 2016: Paper Session 4 - Gaze Prediction for Recommender Systems
0:17:30
RecSys 2016: Paper Session 6 - Domain-Aware Grade Prediction and Top-n Course Recommendation
0:13:47
Session 8: You Say Factorization Machine, I Say Neural Network It’s All in the Activation
0:19:09
RecSys 2016: Paper Session 1 - A Coverage-Based Approach to Recommendation
0:16:41
RecSys 2016: Paper Session 2 - Learning Hierarchical Feature Influence for Recommendation
0:12:15
RecSys 2016: Paper Session 4 - ExpLOD: A Framework for Explaining Recommendations
Комментарии