Physics

Relational model

Biology

Operating system

Algorithm

Optics

Data mining

Epistemology

Philosophy

Artificial intelligence

Focus (optics)

Natural language processing

Class (philosophy)

Sentence

Sequence (biology)

Computer science

Relational database

Mechanism (biology)

Process (computing)

Genetics

Decoding methods

End-to-end principle

The relational facts in sentences are often complicated. Different relational triplets may have overlaps in a sentence. We divided the sentences into three types according to triplet overlap degree, including Normal, EntityPairOverlap and SingleEntiyOverlap. Existing methods mainly focus on Normal class and fail to extract relational triplets precisely. In this paper, we propose an end-to-end model based on sequence-to-sequence learning with copy mechanism, which can jointly extract relational facts from sentences of any of these classes. We adopt two different strategies in decoding process: employing only one united decoder or applying multiple separated decoders. We test our models in two public datasets and our model outperform the baseline method significantly.

Extracting Relational Facts by an End-to-End Neural Model with Copy Mechanism