Papers with Separate Head-Tail Convolution Neural Network
Distant Supervised Relation Extraction with Separate Head-Tail CNN (D19-55)
Copied to clipboard
| Challenge: | Distant supervised relation extraction suffers from mislabeling problem due to strong assumption . et al., 2009: distant supervision can be used to extract relations between entities . |
| Approach: | They propose a neural relation extraction framework which uses convolution and pooling to extract better semantic features of sentences. |
| Outcome: | The proposed framework improves relation extraction performance on a widely used dataset. |