Skip to content

Linearity Trick for Complex Embedding Model Enhancing Link Prediction on Knowledge Graphs

License

Notifications You must be signed in to change notification settings

stmrdus/LinearComplEx

Repository files navigation

LinearComplEx

Linearity for Link Prediction Complex Embedding Model

GitHub top languageGitHub issues GitHub repo size GitHub last commit GitHub forks GitHub stars GitHub

WN18RR Experiment

Model MRR MR Hits@1 Hits@3 Hits@10
DistMult (Yang et al. 2015) 0,43 5110 0,39 0,44 0,49
ComplEx (Trouillon et al. 2016) 0,44 5261 0,41 0,46 0,51
ConvE (Tim Dettmers 2018) 0,45412 5260,6 0,39 0,43 0,48
LinearComplEx (Ours) 0,449 6159 0,422 0,462 0,501

FB15K-237 Experiment

Model MRR MR Hits@1 Hits@3 Hits@10
DistMult (Yang et al. 2015) 0,241 254 0,155 0,263 0,419
ComplEx (Trouillon et al. 2016) 0,247 339 0,158 0,275 0,428
R-GCN (Schlichtkrull et al. 2017) 0,248 0,153 0,258 0,417
ConvE (Tim Dettmers 2018) 0,316 246 0,239 0,35 0,491
LinearComplEx (Ours) 0,307 6159 597 0,339 0,474

Contact

Nhut-Nam Le - @Namlee298 - [email protected]

Project Link: https://github.com/nhutnamhcmus/LinearComplEx

This project is based on https://github.com/ibalazevic/HypER

Acknowledgements

This research is funded by the University of Science, VNU-HCM, Vietnam under grant number CNTT 2022-2 and Advanced Program in Computer Science.

About

Linearity Trick for Complex Embedding Model Enhancing Link Prediction on Knowledge Graphs

Resources

License

Code of conduct

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages