Attention-Based Capsule Networks with Dynamic Routing for Relation Extraction
Ningyu Zhang, Shumin Deng, Zhanling Sun, Xi Chen, Wei Zhang, Huajun Chen · 2018
A capsule is a group of neurons, whose activity vector represents the instantiation parameters of a specific type of entity.In this paper, we explore the capsule networks used for relation extraction in a multi-instance multilabel learning framework and propose a novel neural approach based on capsule networks with attention mechanisms.We evaluate our method with different benchmarks, and it is demonstrated that our method improves the precision of the predicted relations.Particularly, we show that capsule networks improve multiple entity pairs relation extraction 1 .* Corresponding author. 1 In this paper, multiple entity pairs relation extraction refers to multiple entity pairs in a single sentence and each pair of entities contains only one relation label.