Abstract
Most biomedical information extraction efforts are focused on binary relations, there is a strong need to extract drug–gene–mutation n-ary relations among cross-sentences. In recent years, end-to-end biomedical relation extraction with sequence-based or dependency-based method has gained increasing attention. However, handling global dependencies and structural information remains challenges for sequence-based and dependency-based models. Joint exploitation of sequence and graph information may improve biomedical cross-sentence relation extraction. In this paper, we present a hybrid model for extracting biomedical relation in a cross-sentence which aims to address these problems. Our models rely on the self-attention mechanism that directly draws the global dependency relation of the sentence. Furthermore, to preserve the dependency structural information between the words that contain the syntactic dependency relations, we employ graph convolutional networks that encode the dependency structural information to guide the multihead attention to learn the dependency relation. Through extensive experiments on benchmark datasets, we demonstrated the effectiveness of our method.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.