For Neural Relation Extraction (NRE), Recurrent Neural Networks (RNNs) are utilized to handle long sequential data. They can be applied over shortest dependency paths to effectively integrate syntactic information, with deep RNNs making further use of these paths.

Columbia University

Google

- Recursive neural networks
- Convolutional neural networks 
- Recurrent neural networks 
- Graph neural networks 
- Attention-based neural networks


Architectures explored for NRE


CNNs effectively model local textual patterns and can be adopted over shortest dependency paths in order to integrate syntactic information.


Convolutional neural networks for NRE


GNNs that build word/entity graphs for reasoning can be used


Graph neural networks for NRE


Attention-based neural networks for NRE use attention mechanism to aggregate global relational information 


Attention-based neural networks for NRE


Recurrent Neural Networks for Neural Relation Extraction

Recursive neural networks learn compositional representations for sentences recursively. In the context of neural relation extraction (NRE), networks based on augmented dependency paths are utilized to effectively integrate syntactic information.

Learn Before

Related