融合结构与语义信息的知识推理算法
Knowledge Inference Algorithm Based on Combination of Structure and Context
摘要: 知识图谱以三元组形式存储了大量的事实、知识，但同时也存在着事实缺失的问题，因此需要在图谱的已知事实基础上推理预测新的事实即知识推理。传统的知识推理算法只简单利用了知识图谱的结构信息，对知识图谱的信息挖掘不够充分。本文提出了一个融合语义和结构信息的知识推理算法，该算法在利用知识图谱的结构信息的同时，也利用了大规模文本数据中的上下文信息，能够更加准确地表示实体、关系等知识图谱的基本元素。同时针对知识推理模型训练过程中三元组负采样存在的低质量和假阴性问题，我们引入了生成对抗网络来解决这个问题。实验表明，本算法可以实现良好的知识推理效果。
Abstract: Knowledge graph stores a lot of facts and knowledge by triples, but there is also the problem of lack of fact. Therefore, it is necessary to infer and predict new facts, that is, knowledge inference, based on the known facts of the graph. The traditional knowledge inference algorithms only make use of the structure information of the knowledge graph, which is not sufficient to mine the information of the knowledge graph. This paper proposes a knowledge inference algorithm that combines semantic and structural information. It not only uses the structure information of knowledge graph, but also uses the context information in large-scale text data, which can more accurately represent the basic elements of knowledge graph such as entities and relationships. At the same time, because of the problem of low quality and false negativity of negative sampling of triples in the training of knowledge inference model, we introduce the GANs to solve this problem. Experiments show that this algorithm can achieve good knowledge inference effect.
文章引用：刘明坤. 融合结构与语义信息的知识推理算法[J]. 计算机科学与应用, 2022, 12(3): 602-609. https://doi.org/10.12677/CSA.2022.123061

参考文献

[1] Bordes, A., Usunier, N., Garcia-Duran, A., et al. (2013) Translating Embeddings for Modeling Multi-Relational Data. 27th Conference on Neural Information Processing Systems, Lake Tahoe, 5-10 December 2013, 2787-2795.
[2] Wang, Z., Zhang, J., Feng, J. and Chen, Z. (2014) Knowledge Graph Embedding by Translating on Hyperplanes. Proceedings of the 28th AAAI Conference on Artificial Intelligence, Québec City, 27-21 July 2014, 1112-1119.
[3] Yao, L., Mao, C. and Luo, Y. (2019) KG-BERT: BERT for Knowledge Graph Completion. arXiv:1909.03193 http://arxiv.org/abs/1909.03193
[4] Socher, R., Chen, D., Manning, C.D., et al. (2013) Reasoning with Neural Tensor Networks for Knowledge Base Completion. Curran Associates Inc., Red Hook.
[5] Neelakantan, A., Roth, B. and McCallum, A. (2015) Compositional Vector Space Models for Knowledge Base Inference. arXiv:1504.06662.
[6] Graves, A., Wayne, G., Reynolds, M., et al. (2016) Hybrid Computing Using a Neural Net-work with Dynamic External Memory. Nature, 538, 471-476. [Google Scholar] [CrossRef] [PubMed]
[7] Nickel, M., Tresp, V. and Kriegel, H.P. (2011) A Three-Way Model for Collective Learning on Multi-Relational Data. Interna-tional Conference on International Conference on Machine Learning, Bellevue, 28 June-2 July 2011, 809-816.
[8] Mikolov, T., Sutskever, I., Chen, K., Corrado, G. and Dean, J. (2013) Distributed Representations of Words and Phrases and Their Compositionality. Proceedings of the 27th International Conference on Neural Information Processing Systems, Vol. 2, Lake Tahoe, 5-8 December 2013, 3111-3119.
[9] Qin, S., Rao, G., Bin, C., Chang, L., Gu, T. and Xuan, W. (2019) Knowledge Graph Embedding Based on Adaptive Negative Sampling. 2019 International Conference of Pioneering Computer Scientists, Engineers and Educators, Guilin, 20-23 September 2019, 551-563.
[10] Zhang, Y., Yao, Q., Shao, Y. and Chen, L. (2019) NSCaching: Simple and Efficient Negative Sampling for Knowledge Graph Embedding. 2019 IEEE 35th International Conference on Data Engineering (ICDE), Macao (China), 8-11 April 2019, 614-625. [Google Scholar] [CrossRef] 
[11] Xie, R., Liu, Z. and Sun, M. (2018) Does William Shakespeare REALLY Write Hamlet? Knowledge Representation Learning with Confidence. arXiv:1705.03202v2.
[12] Shan, Y., Bu, C., Liu, X., Ji, S. and Li, L. (2018) Confidence-Aware Negative Sampling Method for Noisy Knowledge Graph Embedding. 2018 IEEE International Conference on Big Knowledge (ICBK), Sin-gapore, 17-18 November 2018, 33-40. [Google Scholar] [CrossRef] 
[13] Trouillon, T., Welbl, J., Riedel, S., Gaussier, É. and Bouchard, G. (2016) Complex Embeddings for Simple Link Prediction. Proceedings of the 33rd International Conference on International Conference on Machine Learning, Vol. 48, New York, 19 June-24June, 2016, 2071-2080.
[14] Yang, B., Yih, W.-T., He, X., Gao, J. and Deng, L. (2015) Embedding Entities and Relations for Learning and Inference in Knowledge Bases. 3rd International Conference on Learning Representations, San Diego, 7-9 May 2015, 1-13.
[15] Balažević, I., Allen, C. and Hospedales, T.M. (2019) TuckER: Tensor Factorization for Knowledge Graph Completion. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Pro-cessing and the 9th International Joint Conference on Natural Language Processing, Hong Kong (China), 3-7 Novem-ber 2019, 5185-5194. [Google Scholar] [CrossRef] 
[16] Chami, I., Wolf, A., Juan, D.-C., Sala, F., Ravi, S. and Ré, C. (2020) Low-Dimensional Hyperbolic Knowledge Graph Embeddings. Proceedings of the 58th Annual Meeting of the Associa-tion for Computational Linguistics, Online, 5-10 July 2020, 6901-6914. [Google Scholar] [CrossRef]