摘要
机器阅读理解是自然语言处理领域的核心任务,高考阅读理解自动问答是近年来阅读理解任务中的重要挑战。由于高考题难度较大,同时高考阅读理解问答的数据集较少,导致传统的方法答题效果欠佳。基于此,该文提出一种基于异构图神经网络的答案句抽取模型,将丰富的节点(句子节点、词语节点)和节点之间的关系(框架关系、篇章主题关系)引入图神经网络模型中,问句不仅可以通过中继词语节点与候选句节点进行交互,还可以通过框架语义和篇章主题关系与候选节点进行相互更新。不同类型的语义节点和多维度的语义关系可以帮助模型更好地对信息进行筛选、理解和推理。模型在北京高考语文真题上进行测试,实验结果表明,基于图神经网络的问答模型答题效果优于基线模型,F1值达到了78.08%,验证了该方法的有效性。
The question answering of college entrance examination reading comprehension is an important challenge in reading comprehension task in recent years.This paper proposes a model of answer sentence extraction based on heterogeneous graph neural network.Rich relationships(frame semantics and discourse topic relationships)between nodes(sentences and words)are introduced into the graph neural network.Therefore,questions can interact with candidate answer sentences through both words nodes and frame semantics and discourse topic relationships.The results show that the proposed model outperforms the baseline model with 78.08%Fr value.
作者
杨陟卓
李沫谦
张虎
李茹
YANG Zhizhuo;LI Moqian;ZHANG Hu;LI Ru(School of Computer and Information Technology of Shanxi University,Taiyuan,Shanxi 030006,China;Key Laboratory of Computation Intelligence and Chinese Information Processing of Shanxi University,Taiyuan,Shanxi 030006,China)
出处
《中文信息学报》
CSCD
北大核心
2023年第5期101-111,共11页
Journal of Chinese Information Processing
基金
国家重点研发计划项目(2018YFB1005103)
山西省基础研究计划项目面上基金(20210302123469)
国家自然科学基金(62176145)。
关键词
阅读理解问答
答案句抽取
异构图神经网络
框架语义
篇章主题
reading comprehension QA
answer sentence extraction
heterogeneous graph neural network
frame semantics
discoursetopic
