Position within the page tree

KnowGraphs

Knowledge Graphs at Scale

Knowledge graphs (KGs) are a flexible knowledge representation paradigm intended to allow knowledge to be consumed by humans and machines. KGs are widely regarded as a key enabler for a number of increasingly popular technologies including Web search, question answering, personal assistants and AI across most sectors including Industry 4.0, personalized medicine, legislation, economics and more. KGs are now used by several large companies as a key component of their data products. However, while they are rightly praised as a key technology for all future data-driven enterprises and regarded as a promising approach towards “blurring the lines between human and machine”, KGs are currently unattainable for the majority of companies and users.
The objective of KnowGraphs is to scale KGs to be accessible to a wide audience of users across multiple domains including companies (in domains including Industry 4.0, biomedicine, finance, law) of all sizes and even end users (e.g., through personal assistants and web search). Addressing this goal demands a mix of works from the theoretical foundations to the exploitation and economic repercussions of knowledge graphs.

Operating Time: 10/2019 - 09/2023

Source of Funding: EU Horizon 2020

Partners:

Web Site: KnowGraphs

Publications

Lu, J., Shen, J., Xiong, B., Ma, W., Staab, S., & Yang, C. (2023). HiPrompt: Few-Shot Biomedical Knowledge Fusion via Hierarchy-Oriented Prompting. The 46th International ACM SIGIR Conference on Research and Development in Information Retrieval-Short Paper 2023.
- BibTeX
BibTeX
@conference{lu2023hiprompt, author = {Lu, Jiaying and Shen, Jiaming and Xiong, Bo and Ma, Wenjing and Staab, Steffen and Yang, Carl}, booktitle = {The 46th International ACM SIGIR Conference on Research and Development in Information Retrieval-Short Paper 2023}, title = {HiPrompt: Few-Shot Biomedical Knowledge Fusion via Hierarchy-Oriented Prompting}, year = 2023 }
Gregucci, C., Nayyeri, M., Hernández, D., & Staab, S. (2023). Link Prediction with Attention Applied on Multiple Knowledge Graph Embedding Models.
- BibTeX
BibTeX
@misc{gregucci2023link, archiveprefix = {arXiv}, author = {Gregucci, Cosimo and Nayyeri, Mojtaba and Hernández, Daniel and Staab, Steffen}, eprint = {2302.06229}, primaryclass = {cs.AI}, title = {Link Prediction with Attention Applied on Multiple Knowledge Graph Embedding Models}, year = 2023 }
Xiong, B., Cochez, M., Nayyeri, M., & Staab, S. (2022). Hyperbolic Embedding Inference for Structured Multi-Label Prediction. Advances in Neural Information Processing Systems 2022. https://openreview.net/forum?id=XFnDhcEH9FF
- BibTeX
- Link
BibTeX
@conference{noauthororeditor, author = {Xiong, Bo and Cochez, Michael and Nayyeri, Mojtaba and Staab, Steffen}, booktitle = {Advances in Neural Information Processing Systems 2022}, title = {Hyperbolic Embedding Inference for Structured Multi-Label Prediction}, url = {https://openreview.net/forum?id=XFnDhcEH9FF}, year = 2022 }
Link
https://openreview.net/forum?id=XFnDhcEH9FF
Xiong, B., Potyka, N., Tran, T.-K., Nayyeri, M., & Staab, S. (2022). Faithful Embedding for EL++ Knowledge Bases. Proceedings of the 21st International Semantic Web Conference (ISWC 2022), 1–16. https://arxiv.org/abs/2201.09919
- BibTeX
- Link
BibTeX
@inproceedings{noauthororeditor2022faithful, author = {Xiong, Bo and Potyka, Nico and Tran, T.-K. and Nayyeri, Mojtaba and Staab, Steffen}, booktitle = {Proceedings of the 21st International Semantic Web Conference (ISWC 2022)}, pages = {1-16}, preprinturl = {https://arxiv.org/abs/2201.09919}, title = {Faithful Embedding for EL++ Knowledge Bases}, type = {Publication}, url = {https://arxiv.org/abs/2201.09919}, year = 2022 }
Link
https://arxiv.org/abs/2201.09919
Xiong, B., Zhu, S., Nayyeri, M., Xu, C., Pan, S., Zhou, C., & Staab, S. (2022, August). Ultrahyperbolic Knowledge Graph Embeddings. 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD-22. https://doi.org/10.1145/3534678.3539333
- BibTeX
- Link
BibTeX
@inproceedings{boxiong2022ultrahyperbolic, abstract = {Recent knowledge graph (KG) embeddings have been advanced by hyperbolic geometry due to its superior capability for representing hierarchies. The topological structures of real-world KGs, however, are rather heterogeneous, i.e., a KG is composed of multiple distinct hierarchies and non-hierarchical graph structures. Therefore, a homogeneous (either Euclidean or hyperbolic) geometry is not sufficient for fairly representing such heterogeneous structures. To capture the topological heterogeneity of KGs, we present an ultrahyperbolic KG embedding (UltraE) in an ultrahyperbolic (or pseudo-Riemannian) manifold that seamlessly interleaves hyperbolic and spherical manifolds. In particular, we model each relation as a pseudo-orthogonal transformation that preserves the pseudo-Riemannian bilinear form. The pseudo-orthogonal transformation is decomposed into various operators (i.e., circular rotations, reflections and hyperbolic rotations), allowing for simultaneously modeling heterogeneous structures as well as complex relational patterns. Experimental results on three standard KGs show that UltraE outperforms previous Euclidean, hyperbolic, and mixed-curvature KG embedding approaches.}, author = {Xiong, Bo and Zhu, Shichao and Nayyeri, Mojtaba and Xu, Chengjin and Pan, Shirui and Zhou, Chuan and Staab, Steffen}, booktitle = {28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD-22}, doi = {https://doi.org/10.1145/3534678.3539333}, month = {08}, title = {Ultrahyperbolic Knowledge Graph Embeddings}, url = {https://arxiv.org/abs/2206.00449}, year = 2022 }
Link
https://arxiv.org/abs/2206.00449
Xiong, B., Zhu, S., Potyka, N., Pan, S., Zhou, C., & Staab, S. (2022). Pseudo-Riemannian Graph Convolutional Networks. Advances in Neural Information Processing Systems. https://arxiv.org/abs/2106.03134
- BibTeX
- Link
BibTeX
@inproceedings{xiong2022pseudoriemannian, abstract = {Graph Convolutional Networks (GCNs) are typically studied through the lens of Euclidean geometry. Non-Euclidean Riemannian manifolds provide specific inductive biases for embedding hierarchical or spherical data, but cannot align well with data of mixed topologies. We consider a larger class of semi-Riemannian manifolds with indefinite metric that generalize hyperboloid and sphere as well as their submanifolds. We develop new geodesic tools that allow for extending neural network operations into geodesically disconnected semi-Riemannian manifolds. As a consequence, we derive a principled Semi-Riemannian GCN that first models data in semi-Riemannian manifolds of constant nonzero curvature in the context of graph neural networks. Our method provides a geometric inductive bias that is sufficiently flexible to model mixed heterogeneous topologies like hierarchical graphs with cycles. Empirical results demonstrate that our method outperforms Riemannian counterparts when embedding graphs of complex topologies.}, author = {Xiong, Bo and Zhu, Shichao and Potyka, Nico and Pan, Shirui and Zhou, Chuan and Staab, Steffen}, booktitle = {Advances in Neural Information Processing Systems}, title = {Pseudo-Riemannian Graph Convolutional Networks}, url = {https://arxiv.org/abs/2106.03134}, year = 2022 }
Link
https://arxiv.org/abs/2106.03134

KnowGraphs

Publications

BibTeX

BibTeX

BibTeX

Link

BibTeX

Link

BibTeX

Link

BibTeX

Link

Audience

Formalities

Services

Organization

KnowGraphs

Publications

BibTeX

BibTeX

BibTeX

Link

BibTeX

Link

BibTeX

Link

BibTeX

Link

Here you can reach us

Audience

Formalities

Services

Organization