知识图谱嵌入是一种将实体和关系映射到低维向量空间的技术。目前已有的嵌入表示方法在对具有不对等特征的知识图谱中的实体和关系建模时存在两大缺陷: 一是假定头尾实体来自同一语义空间,忽略二者在链接结构和数量上的不对等;二是每个关系单独配置一个投影矩阵,忽略关系之间的内在联系,导致知识共享困难,泛化能力差。该文提出一种新的嵌入表示方法TransRD,首先对头尾实体采用不对等转换矩阵进行投影,并用ADADELTA算法自适应调整学习率;其次对关系按相关性分组,每组关系使用同一对投影矩阵的方式来共享公共信息,解决泛化能力差的问题。在公开的数据集WN18和FB15K以及MPBC_20(乳腺癌知识图谱的子集)上进行实验和结果分析并与现有的模型进行对比,结果表明TransRD在各项指标上均取得大幅提升。
Abstract
Knowledge graph embedding maps entities and relations into low-dimensional vector spaces. Existing embedding representation methods have two major drawbacks in modeling knowledge graph with asymmetric characteristics. First, they do not consider asymmetry between head and tail entities, assuming that the head and tail entities in knowledge graphs come from the same semantic spaces. Second, they equip each relation with a set of unique projection matrices, ignoring the intrinsic correlations of relations, which hinder the sharing of knowledge between projection matrices and cause poor generalization ability. This paper proposes a novel embedding approach named Trans-RD to deal with the two issues above. TransRD adopts different projection matrices for head and tail entities respectively, and applies ADADELTA algorithm to adjust the learning rate adaptively. Then it uses the same pair of transfer matrices for similar relations to improve the performance of knowledge representation. Empirical results of link prediction based on WN18 and FB15K (public knowledge graph datasets) and MPBC_20 (a subset of Knowledge Graph of Breast Cancer) show that TransRD achieves remarkable improvement in various aspects compared to existing models.
关键词
知识图谱嵌入 /
不对等投影 /
关系相关性
{{custom_keyword}} /
Key words
knowledge graph embedding /
asymmetric mapping /
correlations of relation
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Bordes A,Usunier N,Garcia-Duran A,et al. Translating embeddings for modeling multi-relational data[C]//Proceedings of NIPS 2013. Cambridge,MA: MIT Press,2013: 2787-2795.
[2] Nickel M,Tresp V,Kriegel H P. A three-way model for collective learning on multi-relational data[C]//Proceedings of ICML 2011.New York: ACM,2011: 809-816.
[3] Socher R,Chen D,Manning C D,et al. Reasoning with neural tensor networks for knowledge base completion[C]//Proceedings of NIPS 2013. Cambridge,MA: MIT Press,2013: 926-934.
[4] Trouillon T,Dance C R,Gaussier ,et al. Knowledge graph completion via complex tensor factorization[J]. The Journal of Machine Learning Research,2017,18(1): 4735-4772.
[5] 杜治娟,张祎,孟小峰,等. EAE: 一种酶知识图谱自适应嵌入表示方法[J]. 计算机研究与发展,2017,54(12): 2674-2686.
[6] Ji G,Liu K,He S,et al. Knowledge graph completion with adaptive sparse transfer matrix[C]//Proceedings of AAAI 2016. Menlo Park,CA: AAAI ,2016: 985-991.
[7] Lin Y,Liu Z,Sun M,et al. Learning entity and relation embeddings for knowledge graph completion[C]//Proceedings of AAAI 2015. Menlo Park,CA: AAAI,2015: 2181-2187.
[8] Zhu J Z,Jia Y T,Xu J,et al. Modeling the correlations of relations for knowledge graph embedding[J]. Journal of Computer Science and Technology,2018,33(2): 323-334.
[9] Xie Q,Ma X,Dai Z,et al. An interpretable knowledge transfer model for knowledge base completion[C]//Proceedings of ACL 2017. Stroudsburg,PA: ACL,2017: 950-962.
[10] Nguyen D Q,Sirts K,Qu L,et al. STransE: A novel embedding model of entities and relationships in knowledge bases[C]//Proceedings of NAACL HLT 2016. Stroudsburg,PA: ACL,2016: 460-466.
[11] Zeiler M D. ADADELTA: An adaptive learning rate method[J]. arXiv preprint arXiv: 1212/5701,2012.
[12] Nickel M,Murphy K,Tresp V,et al. A review of relational machine learning for knowledge graphs[C]//Proceedings of the IEEE,2016: 11-33.
[13] Wang Q,Mao Z,Wang B,et al. Knowledge graph embedding: A survey of approaches and applications[J]. IEEE Transactions on Knowledge and Data Engineering,2017,29(12): 2724-2743.
[14] He S,Liu K,Ji G,et al. Learning to represent knowledge graphs with Gaussian embedding[C]//Proceedings of ACM Int. New York: ACM ,2015: 623-632.
[15] Bordes A,Weston J,Collobert R,et al. Learning structured embeddings of knowledge bases[C]//Proceedings of AAAI 2011,Menlo Park,CA: AAAI,2011: 301-306.
[16] Bordes A,Glorot X,Weston J,et al. A semantic matching energy function for learning with multi-relational data[J]. Machine Learning,2014,94(2): 233-259.
[17] 方阳,赵翔,谭真,等. 一种改进的基于翻译的知识图谱表示方法[J]. 计算机研究与发展,2018,55(1): 139-150.
[18] Ji G,He S,Xu L,et al. Knowledge graph embedding via dynamic mapping matrix[C]//Proceedings of ACL 2015. Stroudsburg PA: ACL,2015: 687-696.
[19] Tian F,Gao B,Chen E H,et al. Learning better word embedding by asymmetric low-rank projection of knowledge graph[J]. Journal of Computer Science and Technology,2016,31(3): 624-634.
[20] Wang Z,Zhang J,Feng J,et al. Knowledge graph embedding by translating on hyperplanes[C]//Proceedings of AAAI 2014. Menlo Park,CA: AAAI,2014: 1112-1119.
[21] 段鹏飞,王远,熊盛武,等. 基于空间投影和关系路径的地理知识图谱表示学习[J]. 中文信息学报,2018,32(3): 26-33.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家自然科学基金(71531012)
{{custom_fund}}
PDF(3333 KB)
