基于实体画像增强网络的事件检测方法

事件检测任务旨在从非结构化的文本中自动识别并分类事件触发词。挖掘和表示实体的属性特征(即实体画像)有助于事件检测,其基本原理在于“实体本身的属性往往暗示了其参与的事件类型”(例如,“警察”往往参与“Arrest-Jail”类的事件)。现有研究已利用编码信息实现实体表示,并借此优化事件检测模型。然而,其表示学习过程仅仅纳入局部的句子级语境信息,使得实体画像的信息覆盖率偏低。为此,该文提出基于全局信息和实体交互信息的画像增强方法,其借助图注意力神经网络,不仅在文档级的语境范围内捕捉实体的高注意力背景信息,也同时纳入了局部相关实体的交互信息。特别地,该文开发了基于共现图的注意力遮蔽模型,用于降低噪声信息对实体表示学习过程的干扰。在此基础上,该文联合上述实体画像增强网络、BERT语义编码网络和GAT聚合网络,形成了总体的事件检测模型。该文在通用数据集ACE 2005上进行实验,结果表明实体画像增强方法能够进一步优化事件检测的性能,在触发词分类任务上的F₁值达到76.2%,较基线模型提升了2.2%。

Entity Profile Enhancement Network for Event Detection

Event detection aims to automatically identify and classify event triggers from unstructured texts. Entity representation(or entity profiling)is deemed to be positive to event detection based on the hypothesis that “the features of entities often implies the type of events they participate in” (for example, “police” often participate in “Arrest-Jail” type of events). In this paper, we propose an entity representation enhancement method based on the document-level context and entities interaction. We use the graph attention neural network to capture the document-level information of entities on the background of high attention network, and take into account the local interactive information of other related entities. In particular, an attention mask model based on the entity co-occurrence graph is developed to reduce the noise information. We finally combine the entity representation enhancement network, BERT semantic encoding network, and GAT aggregation network to form the overall event detection model. Experiments on ACE2005 demonstrate the proposed method achieves 76.2% F₁ score in trigger classification task, outperforming the baseline model by 2.2%.