高级检索
    肖梦南, 贺瑞芳, 马劲松. 基于分层潜在语义驱动网络的事件检测[J]. 计算机研究与发展. doi: 10.7544/issn1000-1239.202220447
    引用本文: 肖梦南, 贺瑞芳, 马劲松. 基于分层潜在语义驱动网络的事件检测[J]. 计算机研究与发展. doi: 10.7544/issn1000-1239.202220447
    Xiao Mengnan, He Ruifang, Ma Jinsong. Event Detection Based on Hierarchical Latent Semantic-Driven Network[J]. Journal of Computer Research and Development. doi: 10.7544/issn1000-1239.202220447
    Citation: Xiao Mengnan, He Ruifang, Ma Jinsong. Event Detection Based on Hierarchical Latent Semantic-Driven Network[J]. Journal of Computer Research and Development. doi: 10.7544/issn1000-1239.202220447

    基于分层潜在语义驱动网络的事件检测

    Event Detection Based on Hierarchical Latent Semantic-Driven Network

    • 摘要: 事件检测旨在检测句子中的触发词并将其分类为预定义的事件类型. 如何有效地表示触发词是实现该任务的核心要素. 目前基于表示的方法通过复杂的深度神经网络来学习候选触发词的语义表示,以提升模型性能. 然而,其忽略了:1)受句子语境的影响,同一个触发词会触发不同的事件类型;2)受自然语言表达多样性的影响,不同的触发词会触发同一个事件类型. 受变分自编码器中隐变量及其他NLP任务中分层结构的启发,提出基于分层潜在语义驱动网络的事件检测方法,通过句子和单词的潜在语义信息来辅助缓解以上2个问题. 模型从文本表示空间中分层降维到新的潜在语义空间,探索事件宏微观语境中更本质的影响信息. 首先,通过BERT对候选句子进行编码,得到句子的表示和句子中单词的表示;其次,设计一个双重的潜在语义机制,并采用VAE挖掘句子和单词级潜在语义;最后,从不同粒度的上下文角度,提出采用一个由粗到细的分层结构来充分使用句子和单词的潜在信息,从而提升模型的性能.ACE2005英文语料库上的实验结果表明,该文所提出方法的F1值在事件检测任务上达到了77.9%. 此外,在实验部分对以上2个问题进行了定量分析,证明了本方法的有效性.

       

      Abstract: Event detection aims to detect triggers in sentences and classify them into pre-defined event types. The key factors lie in appropriately representing triggers. Existing representation-based approaches learn the semantic representation of candidate triggers through complex deep neural networks to improve the performance of models. However, they ignore two important problems: 1) affected by sentence context, the same trigger can trigger different event types; 2) due to the diversity of natural language expression, different triggers can trigger the same event type. Inspired by hidden variables in the variational auto-encoder (VAE) and hierarchical structure in other natural language processing (NLP) tasks, we propose a hierarchical latent semantic-driven network (HLSD) for event detection to address the above two problems through latent semantic information of sentences and words. The model reduces the dimension from the text representation space to the new latent semantic space and explores the more essential influence information in the macro and micro context of events. Firstly, we get the representation of a sentence and the words through BERT. Secondly, a dual latent semantic mechanism is designed, and VAE is used to mine the latent semantic information at the sentence and word level. Finally, from the perspective of different granularity contexts, a hierarchical structure from coarse to fine is proposed to make full use of the latent semantic information of sentences and words, to improve the performance of the model. The experimental results on the ACE2005 corpus show that the F1 performance of the proposed method achieves 77.9%. In addition, this paper quantitatively analyzes the above two problems in the experiment, which proves the effectiveness of the method.

       

    /

    返回文章
    返回