Tree-Enhanced Embedding Method (TEM) Algorithm
(Redirected from Tree-Enhanced Embedding Method (TEM))
Jump to navigation
Jump to search
A Tree-Enhanced Embedding Method (TEM) Algorithm is a supervised item recommendation algorithm that ...
- …
- Counter-Example(s):
- See: Deep Recommender Algorithm, KPRN System, Explainable Recommender Algorithm.
References
2020
- (Feng, He, et al., 2020) ⇒ Fuli Feng, Xiangnan He, Hanwang Zhang, and Tat-Seng Chua. (2020). “Cross-GCN: Enhancing Graph Convolutional Network with k-Order Feature Interactions."
- QUOTE: ... We have also implemented the TEM method of Wang et al. (2018). However, despite considerable effort, it failed to produce reasonable results and we believe that the method is especially suited for the case of a large number of attributes. ...
2019b
- "Exploiting Transfer Learning With Attention for In-Domain Top-N Recommendation."
- QUOTE: ... In this model, the attention mechanism is used to implement local activation of the user interest. The TEM (Tree-Enhanced Embedding) model [35] uses a neural attention layer to allocate distribution weights to features of users and items to provide an explainable recommendation. ACF (Attentive Collaborative Filtering) [36] proposes both itemlevel and component-level attention mechanism to learn the implicit feedback in the multimedia recommendation. ...
2019a
- (Wang, 2019) ⇒ Xiang Wang. (2019). “Exploiting Cross-Channel Information for Personalized Recommendation". PhD Thesis.
- QUOTE: ... the pros and cons of embedding-based and tree-based models complement each other, in terms of generalization ability and interpretability. Hence, to build an effective and explainable recommender systems, a natural solution is to combine the two types of models. ...
... tree-enhanced embedding method (TEM) that unifies the strengths of MF for sparse data modeling and GBDTs for cross feature learning. ...
- QUOTE: ... the pros and cons of embedding-based and tree-based models complement each other, in terms of generalization ability and interpretability. Hence, to build an effective and explainable recommender systems, a natural solution is to combine the two types of models. ...
2018
- (Wang, He, et al., 2018) ⇒ Xiang Wang, Xiangnan He, Fuli Feng, Liqiang Nie, and Tat-Seng Chua. (2018). “TEM: Tree-enhanced Embedding Model for Explainable Recommendation.” In: Proceedings of the 2018 World Wide Web Conference (WWW-2018).
- QUOTE: ... we propose a novel solution named Tree-enhanced Embedding Method (TEM), which combines embedding-based methods with decision tree-based approaches. First, we build a gradient boosting decision trees (GBDT) on the side information of users and items to derive effective cross features. We then feed the cross features into an embedding- based model, which is a carefully designed neural attention network that reweights the cross features according to the current prediction. Owing to the explicit cross features extracted by GBDTs and the easy-to-interpret attention network, the overall prediction process is fully transparent and self-explainable. Particularly, to generate reasons for a recommendation, we just need to select the most predictive cross features based on their attention scores. ...