Academy of Military Medical Sciences, Beijing, 100039, China.
BMC Bioinformatics. 2023 Dec 19;24(1):486. doi: 10.1186/s12859-023-05601-9.
Automatic and accurate extraction of diverse biomedical relations from literature is a crucial component of bio-medical text mining. Currently, stacking various classification networks on pre-trained language models to perform fine-tuning is a common framework to end-to-end solve the biomedical relation extraction (BioRE) problem. However, the sequence-based pre-trained language models underutilize the graphical topology of language to some extent. In addition, sequence-oriented deep neural networks have limitations in processing graphical features.
In this paper, we propose a novel method for sentence-level BioRE task, BioEGRE (BioELECTRA and Graph pointer neural net-work for Relation Extraction), aimed at leveraging the linguistic topological features. First, the biomedical literature is preprocessed to retain sentences involving pre-defined entity pairs. Secondly, SciSpaCy is employed to conduct dependency parsing; sentences are modeled as graphs based on the parsing results; BioELECTRA is utilized to generate token-level representations, which are modeled as attributes of nodes in the sentence graphs; a graph pointer neural network layer is employed to select the most relevant multi-hop neighbors to optimize representations; a fully-connected neural network layer is employed to generate the sentence-level representation. Finally, the Softmax function is employed to calculate the probabilities. Our proposed method is evaluated on three BioRE tasks: a multi-class (CHEMPROT) and two binary tasks (GAD and EU-ADR). The results show that our method achieves F1-scores of 79.97% (CHEMPROT), 83.31% (GAD), and 83.51% (EU-ADR), surpassing the performance of existing state-of-the-art models.
The experimental results on 3 biomedical benchmark datasets demonstrate the effectiveness and generalization of BioEGRE, which indicates that linguistic topology and a graph pointer neural network layer explicitly improve performance for BioRE tasks.
从文献中自动、准确地提取多样化的生物医学关系是生物医学文本挖掘的关键组成部分。目前，基于预训练语言模型堆叠各种分类网络进行微调是端到端解决生物医学关系抽取（BioRE）问题的常用框架。然而，基于序列的预训练语言模型在某种程度上未能充分利用语言的图形拓扑结构。此外，面向序列的深度神经网络在处理图形特征方面存在局限性。
在本文中，我们提出了一种新颖的句子级 BioRE 任务方法，即 BioEGRE（BioELECTRA 和图形指针神经网络关系提取），旨在利用语言的拓扑特征。首先，对生物医学文献进行预处理，以保留涉及预定义实体对的句子。其次，使用 SciSpaCy 进行依存句法分析；根据解析结果将句子建模为图形；使用 BioELECTRA 生成令牌级表示，将其建模为句子图形节点的属性；使用图形指针神经网络层选择最相关的多跳邻居进行优化表示；使用全连接神经网络层生成句子级表示。最后，使用 Softmax 函数计算概率。我们的方法在三个 BioRE 任务上进行了评估：多类（CHEMPROT）和两类（GAD 和 EU-ADR）。结果表明，我们的方法在 CHEMPROT、GAD 和 EU-ADR 三个生物医学基准数据集上的 F1 得分分别达到了 79.97%、83.31%和 83.51%，超过了现有最先进模型的性能。
在 3 个生物医学基准数据集上的实验结果表明，BioEGRE 具有有效性和泛化能力，表明语言拓扑结构和图形指针神经网络层可以显著提高 BioRE 任务的性能。
Zotero 插件
