95,318 Hits in 17.9 sec

5* Knowledge Graph Embeddings with Projective Transformations [article]

Mojtaba Nayyeri, Sahar Vahdati, Can Aykul, Jens Lehmann
2021 arXiv   pre-print
To tackle this problem, we propose a novel KGE model (5*E) in projective geometry, which supports multiple simultaneous transformations - specifically inversion, reflection, translation, rotation, and  ...  Performing link prediction using knowledge graph embedding models has become a popular approach for knowledge graph completion.  ...  Acknowledgements We acknowledge the support of the following projects: SPEAKER (BMWi FKZ 01MK20011A), JOSEPH (Fraunhofer Zukunftsstiftung), Cleopatra (GA 812997), the excellence clusters ML2R (BmBF FKZ  ... 
arXiv:2006.04986v2 fatcat:5b5y6gwutzgilchhogtwisitdy

Structure embedding for knowledge base completion and analytics

Zili Zhou, Guandong Xu, Wenhao Zhu, Jinyan Li, Wu Zhang
2017 2017 International Joint Conference on Neural Networks (IJCNN)  
BGNSE model uses one matrix for each relation, the relation transform between two entities can be done directly by forward and backward propagation of bipartite graph network, no need for subspace projection  ...  To measure the truth of one triplet (The knowledge represented by triplet is true or false), some current embedding methods such as Structured Embedding (SE) project entity vectors into subspace, the meaning  ...  We get the top 200 and top 500 entities from FB15k dataset by entity frequency.  ... 
doi:10.1109/ijcnn.2017.7965925 dblp:conf/ijcnn/ZhouXZLZ17 fatcat:jct77xwovrejxpsxwyjiy4wfmq

A Survey on Knowledge Graph Embedding: Approaches, Applications and Benchmarks

Yuanfei Dai, Shiping Wang, Neal N. Xiong, Wenzhong Guo
2020 Electronics  
For alleviating the issue, knowledge graph embedding is proposed to embed entities and relations in a KG to a low-, dense and continuous feature space, and endow the yield model with abilities of knowledge  ...  Finally, we collect several hurdles that need to be overcome and provide a few future research directions for knowledge graph embedding.  ...  82 53.7 79.9 STransE [53] 217 206 80.9 93.4 219 69 51.6 79.7 TransA [23] 405 392 82.3 94.3 155 74 56.1 80.4 KG2E [56] 362 348 80.5 93.2 183 69 47.5 71.5 TransG [59] 357  ... 
doi:10.3390/electronics9050750 fatcat:zr7i5xmb7nfghhpshvgvk4durq

Orthogonal Relation Transforms with Graph Context Modeling for Knowledge Graph Embedding [article]

Yun Tang, Jing Huang, Guangtao Wang, Xiaodong He, Bowen Zhou
2020 arXiv   pre-print
Translational distance-based knowledge graph embedding has shown progressive improvements on the link prediction task, from TransE to the latest state-of-the-art RotatE.  ...  The proposed method includes two-folds, first we extend the RotatE from 2D complex domain to high dimension space with orthogonal transforms to model relations for better modeling capacity.  ...  We fix the entity embedding dimension as 400, and vary the sub-embedding size from 2, 5, 10, 20, 50, all the way to 100. The blue line and green bar represent MRR and H@10 value, respectively.  ... 
arXiv:1911.04910v3 fatcat:iuhb7wxduzbqlkiil3r6w3lpci

Knowledge-enhanced Session-based Recommendation with Temporal Transformer [article]

Rongzhi Zhang, Yulong Gu, Xiaoyu Shen, Hui Su
2021 arXiv   pre-print
In this paper, we propose a framework called Knowledge-enhanced Session-based Recommendation with Temporal Transformer (KSTT) to incorporate such information when learning the item and session embeddings  ...  Specifically, a knowledge graph, which models contexts among items within a session and their corresponding attributes, is proposed to obtain item embeddings through graph representation learning.  ...  53.10 18.25 Baseline (+KGNN, +SAN, +MTE) 72.09 30.63 53.12 18.24 Baseline (+KGNN, +SAN, +T2v) 72.10 30.70 53.15 18.27 Baseline (+KGNN, +SAN, +TBE) 72.25 31.05 53.25 18.29 4 RELATED WORK 4.1 Session-based  ... 
arXiv:2112.08745v1 fatcat:epftrmha7fdkvm5tg2bvotimna

Learning Visual Models using a Knowledge Graph as a Trainer [article]

Sebastian Monka, Lavdim Halilaj, Stefan Schmid, Achim Rettinger
2021 arXiv   pre-print
The auxiliary knowledge is first encoded in a knowledge graph with respective concepts and their relationships, which is then transformed into a dense vector representation via an embedding method.  ...  The results show that a visual model trained with a knowledge graph as a trainer outperforms a model trained with cross-entropy in all experiments, in particular when the domain gap increases.  ...  Acknowledgement This publication was created as part of the research project "KI Delta Learning" (project number: 19A19013D) funded by the Federal Ministry for Economic Affairs and Energy (BMWi) on the  ... 
arXiv:2102.08747v2 fatcat:cwcc7ric5fb35b5lwladfpckpq

Learning Multi-faceted Knowledge Graph Embeddings for Natural Language Processing

Muhao Chen, Carlo Zaniolo
2017 Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence  
Knowledge graphs have challenged the present embedding-based approaches for representing their multifacetedness.  ...  existing monolingual knowledge with important relational properties and hierarchies.  ...  Introduction Knowledge graph (KG) embeddings are the essential tools that transfer the important concepts of the complicated human languages into machine-understandable representations.  ... 
doi:10.24963/ijcai.2017/744 dblp:conf/ijcai/ChenZ17 fatcat:p47ztk7r25awhmebk66tlw535u

COMET: Commonsense Transformers for Automatic Knowledge Graph Construction

Antoine Bosselut, Hannah Rashkin, Maarten Sap, Chaitanya Malaviya, Asli Celikyilmaz, Yejin Choi
2019 Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics  
commonsense knowledge graphs.  ...  Contrary to many conventional KBs that store knowledge with canonical templates, commonsense KBs only store loosely structured open-text descriptions of knowledge.  ...  Acknowledgments We thank Thomas Wolf, Ari Holtzman, Chandra Bhagavatula, Peter Clark, Rob Dalton, Ronan Le Bras, Rowan Zellers and Scott Yih for helpful discussions over the course of this project, as  ... 
doi:10.18653/v1/p19-1470 dblp:conf/acl/BosselutRSMCC19 fatcat:xosmwkdenrerxk7i2sxx43k7aq

On2Vec: Embedding-based Relation Prediction for Ontology Population [chapter]

Muhao Chen, Yingtao Tian, Xuelu Chen, Zijun Xue, Carlo Zaniolo
2018 Proceedings of the 2018 SIAM International Conference on Data Mining  
Recent advances in translation-based graph embedding methods for populating instance-level knowledge graphs lead to promising new approaching for the ontology population problem.  ...  However, unlike instance-level graphs, the majority of relation facts in ontology graphs come with comprehensive semantic relations, which often include the properties of transitivity and symmetry, as  ...  CN30k and YG15k shares the configuration as k = 50, γ 1 = 0.5, λ = 0.001, α 2 = 0.5, and l 2 norm.  ... 
doi:10.1137/1.9781611975321.36 dblp:conf/sdm/ChenTCXZ18 fatcat:mhv6wk64mjforloufbtuysdxei

Multi-view Modeling to Support Embedded Systems Engineering in SysML [chapter]

Aditya A. Shah, Aleksandr A. Kerzhner, Dirk Schaefer, Christiaan J. J. Paredis
2010 Lecture Notes in Computer Science  
Embedded systems engineering problems often involve many domains, each with their own experts and tools.  ...  To maintain consistency between these domain-specific views, model transformations are defined that map the interdependent constructs to and from a common SysML model.  ...  Fig. 5 .Fig. 6 . 56 A schematic for an Electro-hydraulic Log Splitter represented in EPLAN.  ... 
doi:10.1007/978-3-642-17322-6_25 fatcat:4z6c36igh5b7xaoby43ynvxvty

Semantic Relation Reasoning for Shot-Stable Few-Shot Object Detection [article]

Chenchen Zhu, Fangyi Chen, Uzair Ahmed, Zhiqiang Shen, Marios Savvides
2021 arXiv   pre-print
We also identify the problems of trivially using the raw embeddings with a heuristic knowledge graph and propose to augment the embeddings with a dynamic relation graph.  ...  The detector is trained to project the image representations of objects into this embedding space.  ...  56.8 Table 5.  ... 
arXiv:2103.01903v2 fatcat:vluyswfeencnrmdfhsa3f2p5bq

Knowledge Graph Enhanced Event Extraction in Financial Documents [article]

Kaihao Guo, Tianpei Jiang, Haipeng Zhang
2021 arXiv   pre-print
Specifically, for extracting events from Chinese financial announcements, our method outperforms the state-of-the-art method by 5.3% in F1-score.  ...  We propose a first event extraction framework that embeds a knowledge graph through a Graph Neural Network and integrates the embedding with regular features, all at document-level.  ...  It is worth noting that GNN+Transformer, as the bestperforming model with an overall F1-score of 81.7%, surpasses Doc2EDAG by a relatively large margin (5.4% for precision, 5.2% for recall, and 5.3% for  ... 
arXiv:2109.02592v1 fatcat:mocehh2drrh45i7qrod5pjhn3i

Translating Embeddings for Knowledge Graph Completion with Relation Attention Mechanism

Wei Qian, Cong Fu, Yu Zhu, Deng Cai, Xiaofei He
2018 Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence  
Knowledge graph embedding is an essential problem in knowledge extraction. Recently, translation based embedding models (e.g., TransE) have received increasingly attentions.  ...  In this paper, we propose a novel knowledge graph embedding method named TransAt to learn the translation based embedding, relation-related categories of entities and relation-related attention simultaneously  ...  The best configuration: γ = 6, lr rate = 0.01, k = 50, α = 1, c = 10 on WN11 and γ = 2, lr rate = 0.001, k = 200, α = 1, c = 200 on FB13.  ... 
doi:10.24963/ijcai.2018/596 dblp:conf/ijcai/QianFZCH18 fatcat:m2ufweodpzfdznlxjgycnna3g4

Patient and Graph Embeddings for Predictive Diagnosis of Drug Iatrogenesis [chapter]

Lina F. Soualmia, Vincent Lafon, Stéfan J. Darmoni
2021 Studies in Health Technology and Informatics  
The documents' embeddings, the graphs' embeddings of biomedical concepts, and patients' embeddings, all of them semantically enriched with aligned formal ontologies and semantic networks, will constitute  ...  as models founded on transformers, and symbolic artificial intelligence.  ...  Preliminary Results We have already developed a vectorial space trained on EDSaN and generated a hybrid semantic annotator [4, 5] , and document embeddings to create inter-scientific paper similarities  ... 
doi:10.3233/shti210205 pmid:34042611 fatcat:u4vjos5nbnc4bkweuy3yuohflm

Finding Interpretable Concept Spaces in Node Embeddings using Knowledge Bases [article]

Maximilian Idahl, Megha Khosla, Avishek Anand
2019 arXiv   pre-print
We use an external knowledge base that is organized as a taxonomy of human-understandable concepts over entities as a guide to identify subspaces in node embeddings learned from an entity graph derived  ...  We propose a method that given a concept finds a linear transformation to a subspace where the structure of the concept is retained.  ...  Fig. 4 : 4 Fig. 4: Concept Ranking for Albert Einstein Fig. 5 : 5 Fig. 5: Concept Ranking for Donald Trump  ... 
arXiv:1910.05030v1 fatcat:6l5pqdvesjb35c3olxkfnvyd6e
« Previous Showing results 1 — 15 out of 95,318 results