Knowledge Base Completion 周天烁 2017年9月6日
Outline Introduction Approach Outlook Latent feature model Word Embedding Observed feature model Path Ranking Algorithm Markov logic networks Outlook Highlight work Frontier research
Introduction Knowledge graph arises Knowledge base completion Datasets prosper Suffer from incompleteness/ heterogeneity/ ambiguity Knowledge base completion/ instances matching/ entity resolution Knowledge base completion triple classification /link prediction (h,?,t)
observed feature model Approach KBC latent feature model Markov logic networks observed feature model
Approach ——latent feature model General idea entity representation:vector/tensor relation : operations on the entity’s vector space typical models: TransE RESCAL MLP
Approach ——latent feature model TransE—translation on word embedding from word2vec to word embedding
Approach ——latent feature model TransE—translation on word embedding translation on word embedding V(king) - V(man) + V(woman) ≈ V(queue) h+r=?t (head,relation,tail)
Approach ——latent feature model other models fijk=eJiWkej
Approach ——latent feature model pros powerful Scalability cons low interpretability dependency on hyperparameter
Approach ——observed feature model intuition similar entities are likely to be related (homophily) the similarity of entities can be derived from the paths between nodes typical models: Inductive Logic Programming (ILP) Path Ranking Algorithm(PRA)
Approach ——observed feature model Path Ranking Algorithm(PRA)
Approach ——observed feature model Path Ranking Algorithm(PRA)
Approach ——observed feature model Path Ranking Algorithm(PRA)
Approach ——observed feature model Path Ranking Algorithm(PRA) pros high interpretability improvement space cons Scalability efficiency
Approach ——markov logic network Marckov Random field first order logic markov logic network
Approach ——markov logic network pros suitable for good structure; e.g. NLP(chain),CV(grid); GIS(coordinate); cons computational intractable rule leaning
Outlook highlight frontier research related academic area word embedding PRA rule leaning frontier research hybrid model DeepLearning for word embeddings PRA refinement common sense knowledge reasoning search personalization related academic area statistical relational learning(SRL) link prediction in other networks
Reference [Dat Quoc Nguyen,2017] An overview of embedding models of entities and relationships for knowledge base completion. Yang, Fan, Zhilin Yang, and William W. Cohen. "Differentiable Learning of Logical Rules for Knowledge Base Completion." arXiv preprint arXiv:1702.08367 (2017). Lin, Xixun, Yanchun Liang, and Renchu Guan. "Compositional learning of relation paths embedding for knowledge base completion." arXiv preprint arXiv:1611.07232 (2016). [Bordes et al., 2011] Learning structured embeddings of knowledge bases [Bordes et al., 2013] Translating embeddings for modeling multi-relational data. [Nickel et al., 2015] A review of relational machine learning for knowledge graphs. [Lao and Cohen, 2010] Relational retrieval using a path-constrained random walks. [Lao et al., 2011] Random walk inference and learning in a large scale knowledge base. M. Richardson and P. Domingos, “Markov logic networks,” Machine Learning, vol. 62, no. 1, pp. 107–136, 2006. Wang, Quan, Bin Wang, and Li Guo. "Knowledge Base Completion Using Embeddings and Rules." IJCAI. 2015. Gardner, Matt, and Tom M. Mitchell. "Efficient and Expressive Knowledge Base Completion Using Subgraph Feature Extraction." EMNLP. 2015. zhouzhihua. Machine Learning. 2016 http://www.cnblogs.com/chenbjin/p/5644457.html http://blog.csdn.net/u011274209/article/details/50991385 https://mp.weixin.qq.com/s?__biz=MzIxNzE2MTM4OA==&mid=413465972&idx=2&sn=6e773699812580627 5c5eb1484a8582f&utm_source=tuicool&utm_medium=referral https://wenku.baidu.com/view/ddca21030166f5335a8102d276a20029bd64636e.html http://snap.stanford.edu/data/index.html
Q & A Thanks for listening