Abstract:Background Automatically extracting relations between chemicals and diseases plays an important role in biomedical text mining. Chemical-disease relation (CDR) extraction aims at extracting complex semantic relationships between entities in documents, which contain intrasentence and intersentence relations. Most previous methods did not consider dependency syntactic information across the sentences, which are very valuable for the relations extraction task, in particular, for extracting the intersentence relations accurately. Objective In this paper, we propose a novel end-to-end neural network based on the graph convolutional network (GCN) and multihead attention, which makes use of the dependency syntactic information across the sentences to improve CDR extraction task. Methods To improve the performance of intersentence relation extraction, we constructed a document-level dependency graph to capture the dependency syntactic information across sentences. GCN is applied to capture the feature representation of the document-level dependency graph. The multihead attention mechanism is employed to learn the relatively important context features from different semantic subspaces. To enhance the input representation, the deep context representation is used in our model instead of traditional word embedding. Results We evaluate our method on CDR corpus. The experimental results show that our method achieves an F-measure of 63.5%, which is superior to other state-of-the-art methods. In the intrasentence level, our method achieves a precision, recall, and F-measure of 59.1%, 81.5%, and 68.5%, respectively. In the intersentence level, our method achieves a precision, recall, and F-measure of 47.8%, 52.2%, and 49.9%, respectively. Conclusions The GCN model can effectively exploit the across sentence dependency information to improve the performance of intersentence CDR extraction. Both the deep context representation and multihead attention are helpful in the CDR extraction task.

Document-level Relation Extraction Based on Heterogeneous Graph Convolutional Network and Local Semantic Fusion

Dual-Channel and Hierarchical Graph Convolutional Networks for document-level relation extraction

Document-level relation extraction with multi-layer heterogeneous graph attention network

Document-level Relation Extraction as Semantic Segmentation

Global-to-Local Neural Networks for Document-Level Relation Extraction

Enhanced Heterogeneous Graph Attention Network with a Novel Multilabel Focal Loss for Document-Level Relation Extraction

Dual-stream dynamic graph structure network for document-level relation extraction

Document-level Relation Extraction with Cross-sentence Reasoning Graph

Document-level relation extraction with two-stage dynamic graph attention networks

Reasoning with Latent Structure Refinement for Document-Level Relation Extraction

Long-tail Relation Extraction via Knowledge Graph Embeddings and Graph Convolution Networks.

Local-to-global GCN with knowledge-aware representation for distantly supervised relation extraction

ConstGCN: Constrained Transmission-based Graph Convolutional Networks for Document-level Relation Extraction

A Feature Combination-Based Graph Convolutional Neural Network Model for Relation Extraction

A region-based hypergraph network for joint entity-relation extraction

Double Graph Based Reasoning for Document-level Relation Extraction

HIN: Hierarchical Inference Network for Document-Level Relation Extraction

Document-Level Relation Extraction with Adaptive Thresholding and Localized Context Pooling

Document-Level Biomedical Relation Extraction Using Graph Convolutional Network and Multihead Attention: Algorithm Development and Validation

Graph Enhanced Dual Attention Network for Document-Level Relation Extraction.

Document-Level Relation Extraction with Relation Correlation Enhancement