KGxBoard

Explainable and Interactive Leaderboard for Evaluation of Knowledge Graph Completion Models

Resources

• Video demo: https://vimeo.com/735226047
• Demo website: https://explainaboard.inspiredco.ai/
• Tutorial for the KGxBoard's API and CLI: https://github.com/neulab/ExplainaBoard/blob/main/docs/task_kg_link_tail_prediction.md
  • Note: before using KGxBoard's API or CLI, make sure you've installed ExplainaBoard first: https://github.com/neulab/ExplainaBoard
• LibKGE API for KGxBoard: https://github.com/rufex2001/libkge-with-kgxboard-support
• PyKEEN API for KGxBoard: coming soon

Tutorial for the web interface

1. Go to https://explainaboard.inspiredco.ai/
2. Click on Systems on the left side
3. From the tasks drop-down menu, select kg-link-tail-prediction
4. Click one (or several) of the models and click on the button "Analysis"