文献类型：期刊文献

中文题名：融合图注意力的复杂时序知识图谱推理问答模型

英文题名：Reasoning question answering model of complex temporal knowledge graph with graph attention

作者：蒋汶娟[1];过弋[1,2,3];付娇娇[1]

机构：[1]华东理工大学信息科学与工程学院,上海200237;[2]大数据流通与交易技术国家工程实验室商业智能与可视化技术研究中心,上海200436;[3]上海大数据与互联网受众工程技术研究中心,上海200072

年份：2024

卷号：44

期号：10

起止页码：3047

中文期刊名：计算机应用

外文期刊名：journal of Computer Applications

收录：CSTPCD;;北大核心:【北大核心2023】；CSCD:【CSCD_E2023_2024】；

基金：上海市科学技术委员会科技计划项目(22511104800,22DZ1204903)。

语种：中文

中文关键词：时序知识图谱;复杂问答;图注意力网络;时序推理;时序关系表示

外文关键词：Temporal Knowledge Graph(TKG);complex question answering;Graph Attention neTwork(GAT);temporal reasoning;temporal relationship representation

摘要：在时序知识图谱问答(TKGQA)任务中,针对模型难以捕获并利用问句中隐含的时间信息增强模型的复杂问题推理能力的问题,提出一种融合图注意力的时序知识图谱推理问答(GACTR)模型。所提模型采用四元组形式的时序知识库(KB)进行预训练,同时引入图注意力网络(GAT)以有效捕获问句中隐式时间信息;通过与RoBERTa(Robustly optimized Bidirectional Encoder Representations from Transformers pretraining approach)模型训练的关系表示进行集成,进一步增强问句的时序关系表示;将该表示与预训练的时序知识图谱(TKG)嵌入相结合,以获得最高评分的实体或时间戳作为答案预测结果。在最大的基准数据集CRONQUESTIONS上的实验结果显示,GACTR模型在时序推理模式下能更好地捕获隐含时间信息,有效提升模型的复杂推理能力。与基线模型CRONKGQA(Knowledge Graph Question Answering on CRONQUESTIONS)相比,GACTR模型在处理复杂问题类型和时间答案类型上的Hits@1结果分别提升了34.6、13.2个百分点;与TempoQR(Temporal Question Reasoning)模型相比,分别提升了8.3、2.8个百分点。
In the task of Temporal Knowledge Graph Question Answering(TKGQA),it is a challenge for models to capture and utilize the implicit temporal information in the questions to enhance the complex reasoning ability of the models.To address this problem,a Graph Attention mechanism-integrated Complex Temporal knowledge graph Reasoning question answering(GACTR)model was proposed.The proposed model was pretrained on a temporal Knowledge Base(KB)in the form of quadruples,and a Graph Attention neTwork(GAT)was introduced to effectively capture implicit temporal information in the question.The relationship representation trained by Robustly optimized Bidirectional Encoder Representations from Transformers pretraining approach(RoBERTa)was integrated to enhance the temporal relationship representation of the question.This representation was combined with the pretrained Temporal Knowledge Graph(TKG)embedding,and the final prediction result was the entity or timestamp with the highest score.On the largest benchmark dataset CRONQUESTIONS,compared to the baseline models,Knowledge Graph Question Answering on CRONQUESTIONS(CRONKGQA),the GACTR model achieved improvements of 34.6 and 13.2 percentage points in handling complex question and time answer types,respectively;compared to the Temporal Question Reasoning(TempoQR)model,the improvements were 8.3 and 2.8 percentage points,respectively.