面向脚本事件预测的稠密事件图嵌入

Embedding Dense Event Graph for Script Event Prediction

脚本事件预测是指在给定现有上下文事件链的情况下预测后续事件.在现实世界中,不同事件的关系可以自然地表示为图结构,以事件为节点,以时间或因果关系为边.由于语料库规模有限和信息提取工具的能力不足,先前工作中自动构建的事件图会存在稀疏性问题,并且无法集成来自高阶节点的信息以支持多步推理.为了解决这个问题,本文提出使用可学习的多维加权邻接矩阵的稠密事件图(DEG)来解决之前事件图存在的稀疏性问题并表征事件之间的关系强度.为了实现DEG的嵌入表示,本文同时提出了一个通用框架,该框架能够将高阶事件演化信息组合到事件表示中.在多选叙事完形填空(multiple choice narrative cloze,MCNC)和连贯多选叙事完形填空(coherent multiple choice narrative cloze,CMCNC)数据集上进行了实验,结果证明了此框架的有效性.

Script Event Prediction refers to predicting the subsequent event based on a given existing chain of context events.In the real world,the relationship of different events can be naturally represented as a graph structure,where events serve as nodes and their temporal or causal relations are depicted as edges.However,previous approaches that automatically constructed event graphs suffer from sparsity problem due to the limited scale of corpus and the incapability of information extraction tools.Moreover,they fail to integrate information from higher order nodes to support multi-step reasoning.To remedy this,we propose a Dense Event Graph(DEG)approach which use a learnable multi-dimensional weighted adjacency matrix to address the sparsity issue and characterize the relation strengths between events.To embed the DEG,we propose a general framework capable of combining high-order event evolution information into the event representations. Experimental results on the multiple choicenarrative cloze( MCNC) and coherent multiple choice narrative cloze( CMCNC) demonstrate the effectiveness of ourapproach.