2025年 02期

摘要(Abstract)：

针对传统知识图谱推理模型在时间关联推理方面的局限性，以及现有模型仅通过在静态知识图谱中加入时间戳组合，而未充分考虑时间序列依赖关系的问题，提出基于图表示学习的知识图谱时序推理(KGTR＿GRL)模型；针对图表示学习中的邻居结构信息，设计多关系图结构编码器，以解决当前大部分研究忽略的节点重要性问题；为了更深入地捕获时间信息，将注意力机制引入到时序编码器中，设计模型时序推理算法，通过解码器计算评分并转换为候选实体的概率；采用2个公开数据集测试模型的性能，并与多个现有模型的性能进行比较。结果表明，KGTR＿GRL模型表现出更好的性能，实验中平均倒数排名，预测排名小于或等于1、 10的三元组的平均占比指标均优于其他现有模型，证明了考虑多阶邻居特征信息的多关系编码器性能的优越性。

关键词(KeyWords)：时序推理;时序知识图谱;图表示学习;图卷积神经网络

基金项目(Foundation):国家自然科学基金项目(32060157)

作者(Author): 张宇姣,徐健,吴迪

DOI: 10.13349/j.cnki.jdxbn.20250212.001

参考文献(References)：

[1] BORDES A,USUNIER N,GARCIA-DURáN A,et al.Translating embeddings for modeling multi-relational data[J].Advances in Neural Information Processing Systems,2013,26:2787.

[2] CHEN X J,JIA S B,XIANG Y.A review:knowledge reasoning over knowledge graph[J].Expert Systems with Applications,2020,141:112948.

[3] 夏毅，兰明敬，陈晓慧，等.可解释的知识图谱推理方法综述[J].网络与信息安全学报，2022,8(5):1.

[4] 封皓君，段立，张碧莹.面向知识图谱的知识推理综述[J].计算机系统应用，2021,30(10):21.

[5] BAKHSHI M,NEMATBAKHSH M,MOHSENZADEH M,et al.SParseQA:sequential word reordering and parsing for answering complex natural language questions over knowledge graphs[J].Knowledge-based Systems,2022,235:107626.

[6] 苏瑜鸿.知识图谱中的知识推理技术研究[D].北京：北京邮电大学，2023:41-44.

[7] QIAO C,HU X.A neural knowledge graph evaluator:combining structural and semantic evidence of knowledge graphs for predicting supportive knowledge in scientific QA[J].Information Processing & Management,2020,57(6):102309.

[8] LECUE F.On the role of knowledge graphs in explainable AI[J].Semantic Web,2020,11(1):41.

[9] WANG Q,HAO Y S,CAO J.ADRL:an attention-based deep reinforcement learning framework for knowledge graph reasoning[J].Knowledge-based Systems,2020,197:105910.

[10] ZHANG J S,LIANG S,SHENG Y P,et al.Temporal know-ledge graph representation learning with local and global evolutions[J].Knowledge-based Systems,2022,251:109234.

[11] XIE Z W,ZHU R J,LIU J,et al.A time-aware relational graph attention model for temporal knowledge graph embedding[J].IEEE/ACM Transactions on Audio,Speech,and Language Processing,2023,31:2246.

[12] ZHANG Q X,WENG X Y,ZHOU G Y,et al.ARL:an adaptive reinforcement learning framework for complex question answering over knowledge base[J].Information Processing & Management,2022,59(3):102933.

[13] ZHANG L S,KANG Z,SUN X X,et al.KCRec:knowledge-aware representation graph convolutional network for recommendation[J].Knowledge-based Systems,2021,230:107399.

[14] 姚思雨，赵天哲，王瑞杰，等.规则引导的知识图谱联合嵌入方法[J].计算机研究与发展，2020,57(12):2514.

[15] CHEN L,TANG X,CHEN W Q,et al.Dacha:a dual graph convolution based temporal knowledge graph representation learning method using historical relation[J].ACM Transactions on Knowledge Discovery from Data,2021,16(3):1.

[16] WANG S S,FU K,SUN X,et al.Hierarchical-aware relation rotational knowledge graph embedding for link prediction[J].Neurocomputing,2021,458:259.

[17] XIAO Y,ZHOU G Y,XIE Z W,et al.Learning dual disentangled representation with self-supervision for temporal knowledge graph reasoning[J].Information Processing & Management,2024,61(3):103618.

[18] ZHANG D,FENG W L,WU Z H,et al.CDRGN-SDE:cross-dimensional recurrent graph network with neural stochastic differential equation for temporal knowledge graph embedding[J].Expert Systems with Applications,2024,247:123295.

[19] JIA W,WANG X,SHAN J,et al.Sequence encoder-based spatiotemporal knowledge graph completion[J].Journal of Web Engineering,2022,21(6):1913.