Skip to content

Latest commit

 

History

History
42 lines (26 loc) · 1.43 KB

README.md

File metadata and controls

42 lines (26 loc) · 1.43 KB
Raw

License: MIT

Symbolic Learning and KG Embeddings with Link Prediction and Community Detection

The objective of this workshop is to highlight the impact of symbolic learning over various downstream KG embeddings tasks such as Link Prediction (LP) and Community Detection. For reproducibility, follow the instructions provided in the slides. If running locally, then follow the steps below.

Symbolic Learning

Follow the instructions to execute symbolic learning to generate Enriched, and Transformed KG.

{
  "KG": "OriginalKG",
  "prefix": "http://example.org/lungCancer/entity/",
  "rules_file": "LungCancer-rules-short.csv",
  "rdf_file": "LungCancer.nt",
  "constraints_folder": "Constraints"
}
python symbolic_predictions.py

Knowledge Graph Embeddings and Link Prediction

python SymbolicLearning_KGE/KGEmbedding/kge.py --dataset_path "SymbolicLearning_KGE/KG/OriginalKG/LungCancer.tsv" --output_dir "SymbolicLearning_KGE/KGEmbedding/OriginalKG" --results_path "SymbolicLearning_KGE/KGEmbedding/OriginalKG/" --models TransH

Community Detection

License

This work is licensed under the MIT license.

Authors

Tutorial has been implemented in joint work by Yashrajsinh Chudasama, Disha Purohit, and Ariam Rivas.