基于语义和结构增强的时序知识图谱问答方法

doi:10.3969/j.issn.1006-2475.2024.03.003

摘要/Abstract

摘要： 摘要：知识图谱作为自然语言处理领域中的热门研究主题之一，一直受到学术界的广泛关注。在现实中，知识问答过程往往携带时间信息，因此，近年来，应用时序知识图谱来完成知识问答的技术广泛受到学者的青睐。传统的时序知识图谱问答技术主要通过对问题进行编码来完成推理过程，但其无法处理问题中包含的复杂的实体和时间关系。基于此，提出一种基于语义和结构增强的时序知识图谱问答方法，在推理过程中兼顾问题的语义信息和结构信息，提升对复杂问题正确回答的概率。首先，该方法解析出问题中的隐式时间表达，并基于时序知识图谱中的信息，用直接表达方式改写问题，再根据问题集合中的时间粒度，按照不同的时间粒度聚合时序知识图谱中的时间信息。其次，基于问题中的实体信息和时间信息，对问题语义信息进行表示和融合，以加强对于实体和时间语义的学习。再次，基于提取到的实体完成子图提取，并利用图卷积神经网络提取子图的结构信息。最后，将融合后的问题语义信息与结构信息进行拼接，并对候选答案进行评分，选取评分最高的实体作为答案。在MultiTQ数据集上进行对比测试，实验结果表明，提出的模型优于其他基准模型。

关键词: 关键词：语义增强, 结构增强, 图神经网络, 时序知识图谱问答

Abstract: Abstract： Knowledge graphs， as one of the popular research topics in the field of natural language processing， have consistently received widespread attention from the academic community. In reality， the knowledge quiz process often carries temporal information. Consequently， in recent years， the application of temporal knowledge graphs for knowledge question answering has gained popularity among scholars. Traditional methods for temporal knowledge graph question answering primarily encode the question information to facilitate the inference process. However， they are unable to deal with the more complex entities and temporal relationships contained in the questions. To address this， semantic and structural enhancement for temporal knowledge graph question answering is proposed. This method aims to simultaneously consider both semantic and structural information in the inference process to improve the probability of providing correct answers. Firstly， implicit temporal expressions in the questions are parsed， and the questions are rewritten using direct representations based on the information in the temporal knowledge graph. Additionally， the temporal information in the temporal knowledge graph is aggregated according to different time granularities based on the question set. Secondly， the semantic information of the questions is represented and fused based on entity and time information to enhance the learning of entity and time semantics. Subsequently， subgraphs are extracted based on the extracted entities， and the structural information of the subgraphs is captured using graph convolutional networks. Finally， the fused semantic and structural information of the questions are concatenated， and candidate answers are scored， with the entity receiving the highest score selected as the answer. Comparative tests on MultiTQ data sets show that the proposed model outperforms other baseline models.

Key words: Key words： semantic enhancement, structural enhancement, graph neural networks, temporal knowledge graph question answering

中图分类号:

TP391

黄政霖, 董宝良. 基于语义和结构增强的时序知识图谱问答方法[J]. 计算机与现代化, 2024, 0(03): 15-23.

HUANG Zheng-lin, DONG Bao-liang. Temporal Knowledge Graph Question Answering Method Based on#br# Semantic and Structural Enhancement[J]. Computer and Modernization, 2024, 0(03): 15-23.

参考文献

［1］ SAXENA A， CHAKRABARTI S， TALUKDAR P. Question answering over temporal knowledge graphs［J］. arXiv preprint arXiv：2106.01515， 2021.
［2］ CHEN Z Y， LIAO J， ZHAO X. Multi-granularity temporal question answering over knowledge graphs［C］// Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics. 2023：11378-11392.
［3］ SONG X T， BAI L Y， LIU R K， et al. Temporal knowledge graph entity alignment via representation learning［C］// The 27th International Conference on Database Systems for Advanced Applications. 2022：11-14.
［4］ RADSTOK W， CHEKOL M， VELEGRAKIS Y. Leveraging static models for link prediction in temporal knowledge graphs［C］// 2021 IEEE 33rd International Conference on Tools with Artificial Intelligence （ICTAI）. 2021：1034-1041.
［5］ LIAO S Y， LIANG S S， MENG Z Q， et al. Learning dynamic embeddings for temporal knowledge graphs［C］// Proceedings of the 14th ACM International Conference on Web Search and Data Mining. 2021：535-543.
［6］ LIU K Z， ZHANG Y H. A temporal knowledge graph completion method based on balanced timestamp distribution［J］. arXiv preprint arXiv：2108.13024， 2021.
［7］ WU J P， XU Y S， ZHANG Y X， et al. Tie： A framework for embedding-based incremental temporal knowledge graph completion［C］// Proceedings of the 44th international ACM SIGIR Conference on Research and Development in Information Retrieval. 2021：428-437.
［8］ CHEN K， LI C C， LI A P， et al. Focus on inherent attributes for temporal knowledge graph completion［C］// 2021 International Joint Conference on Neural Networks （IJCNN）.
2021：1-8.
［9］ WANG Y L， XU X Y. ERGCN： Enhanced relational graph convolution network， an optimization for entity prediction tasks on temporal knowledge graphs［J］. Future Internet， 2022，14（12）：376.
［10］ BAI L Y， MA X N， ZHANG M C， et al. TPmod： A tendency-guided prediction model for temporal knowledge graph completion［J］. ACM Transactions on Knowledge Discovery from Data， 2021，15（3）：1-17.
［11］ MA S X， LI A P， ZHAO X J， et al. Learning BiLSTM-based embeddings for relation prediction in temporal knowledge graph［J］. Journal of Physics： Conference Series， 2021，1871（1）：012050.
［12］ LIU Y， HUA W， XIN K X， et al. Context-aware temporal knowledge graph embedding［C］// Web Information Systems Engineering–WISE 2019. 2019：583-598.
［13］ QIAO Y W， SUN L L， XIAO C J. High order semantic relations-based temporal recommendation model by collaborative knowledge graph learning［C］// Asia-Pacific Web （APWeb） and Web-age Information Management （WAIM） Joint International Conference on Web and Big Data. 2020：337-351.
［14］ XU C， CHEN Y Y， NAYYERI M， et al. Temporal knowledge graph completion using a linear temporal regularizer and multivector embeddings［C］// Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics： Human Language Technologies. 2021：2569-2578.
［15］ YU M， GUO J J， YU J， et al. TBDRI： Block decomposition based on relational interaction for temporal knowledge graph completion［J］. Applied Intelligence， 2022，53（5）：5072-5084.
［16］ JIANG T S， LIU T Y， GE T. Towards time-aware knowledge graph completion［C］// COLING 2016， the 26th International Conference on Computational Linguistics. 2016：1715-1724.
［17］ XU C J， NAYYERI M， ALKHOURY F， et al. Temporal knowledge graph completion based on time series gaussian embedding［J］. arXiv preprint arXiv：1911.07893， 2020.
［18］ HU S M， WANG B Y， WANG J P， et al. Transformer-based temporal knowledge graph completion［C］// 2023 IEEE 3rd International Conference on Computer Communication and Artificial Intelligence （CCAI）. 2023.
［19］ CHEN X J， JIA S B， DING L， et al. Reasoning over temporal knowledge graph with temporal consistency constraints［J］. Journal of Intelligent & Fuzzy Systems， 2021，40（6）：11941-11950.
［20］李凤英，范伟豪. 基于时序感知的动态知识图谱补全方法［J］. 计算机工程与应用， 2022，58（15）：202-209.
［21］ MAVROMATIS C， SUBRAMANYAM P L， IOANNIDIS V N， et al. TempoQR： Temporal question reasoning over knowledge graphs［J］. arXiv preprint arXiv：2112.05785， 2021.
［22］ OTTE K， VESTERMARK K S， LI H， et al. Towards a question answering system over temporal knowledge graph embeddings［M］// Deep Learning for Knowledge Graphs 2022. 2022：1-10.
［23］ LACROIX T， OBOZINSKI G， USUNIER N. Tensor decompositions for temporal knowledge base completion［J］. arXiv preprint arXiv：2004.04926， 2020.
［24］ FÉVRY T， SOARES L B， FITZGERALD N， et al. Entities as experts： Sparse memory access with entity supervision［J］. arXiv preprint arXiv：2004.07202， 2020.
［25］ TITOV I， WELLING M， SCHLICHTKRULL M， et al. Modeling relational data with graph convolutional networks［J］. arXiv preprint arXiv：1703.06103， 2017.
［26］ DEVLIN J， CHANG M W， LEE K， et al. BERT： Pre-training of deep bidirectional transformers for language understanding［J］. arXiv preprint arXiv：1810.04805， 2018.
［27］ LAN Z Z， CHEN M D， GOODMAN S， et al. ALBERT： A lite BERT for self-supervised learning of language representations［C］// International Conference on Learning Representations. arXiv preprint arXiv：1909.11942， 2019.
［28］ SANH V， DEBUT L， CHAUMOND J， et al. DistilBERT， a distilled version of BERT： Smaller， faster， cheaper and lighter［J］. arXiv preprint arXiv：1910.01108， 2019.
［29］ SAXENA A， TRIPATHI A， TALUKDAR P. Improving multi-hop question answering over knowledge graphs using knowledge base embeddings［J］. 2020. DOI： 10.18653/v1/2020.acl-main.412.

[1]	马钰, 杨勇, 任鸽, 帕力旦·吐尔逊. 基于GCN和微调BERT的作文自动评分方法[J]. 计算机与现代化, 2024, 0(09): 33-37.
[2]	陈俊义. 基于图节点动静态特征的健康事件预测模型[J]. 计算机与现代化, 2023, 0(10): 39-44.
[3]	夏义春, 李汪根, 李豆豆, 葛英奎, 王志格. 结合注意力机制和图神经网络的CTR预估模型[J]. 计算机与现代化, 2023, 0(03): 29-37.
[4]	姚春华, 张学磊, 宋馨宇, 张举, 蔡佳志, 冯翱. 一种基于图卷积神经网络和依存分析的财经新闻情感分析方法[J]. 计算机与现代化, 2022, 0(05): 33-39.