https://github.com/pminervini/KGReasoning

Multi-Hop Logical Reasoning in Knowledge Graphs
https://github.com/pminervini/KGReasoning

Last synced: about 2 months ago
JSON representation

Multi-Hop Logical Reasoning in Knowledge Graphs

• Host: GitHub
• URL: https://github.com/pminervini/KGReasoning
• Owner: pminervini
• License: mit
• Fork: true (snap-stanford/KGReasoning)
• Created: 2021-03-19T08:50:40.000Z (over 3 years ago)
• Default Branch: main
• Last Pushed: 2022-03-27T20:05:17.000Z (over 2 years ago)
• Last Synced: 2024-07-27T15:43:31.790Z (about 2 months ago)
• Language: Python
• Homepage:
• Size: 41 KB
• Stars: 21
• Watchers: 1
• Forks: 4
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-logical-query - CQD

README

        
# KGReasoning

This repo contains several algorithms for multi-hop reasoning on knowledge graphs, including the official PyTorch implementation of [Beta Embeddings for Multi-Hop Logical Reasoning in Knowledge Graphs](https://arxiv.org/abs/2010.11465) and a PyTorch implementation of [Complex Query Answering with Neural Link Predictors](https://arxiv.org/abs/2011.03459).

**Models**

- [x] [CQD](https://arxiv.org/abs/2011.03459)

- [x] [BetaE](https://arxiv.org/abs/2010.11465)

- [x] [Query2box](https://arxiv.org/abs/2002.05969)

- [x] [GQE](https://arxiv.org/abs/1806.01445)

**KG Data**

The KG data (FB15k, FB15k-237, NELL995) mentioned in the BetaE paper and the Query2box paper can be downloaded [here](http://snap.stanford.edu/betae/KG_data.zip). Note the two use the same training queries, but the difference is that the valid/test queries in BetaE paper have a maximum number of answers, making it more realistic.

Each folder in the data represents a KG, including the following files.

- `train.txt/valid.txt/test.txt`: KG edges

- `id2rel/rel2id/ent2id/id2ent.pkl`: KG entity relation dicts

- `train-queries/valid-queries/test-queries.pkl`: `defaultdict(set)`, each key represents a query structure, and the value represents the instantiated queries

- `train-answers.pkl`: `defaultdict(set)`, each key represents a query, and the value represents the answers obtained in the training graph (edges in `train.txt`)

- `valid-easy-answers/test-easy-answers.pkl`: `defaultdict(set)`, each key represents a query, and the value represents the answers obtained in the training graph (edges in `train.txt`) / valid graph (edges in `train.txt`+`valid.txt`)

- `valid-hard-answers/test-hard-answers.pkl`: `defaultdict(set)`, each key represents a query, and the value represents the **additional** answers obtained in the validation graph (edges in `train.txt`+`valid.txt`) / test graph (edges in `train.txt`+`valid.txt`+`test.txt`)

We represent the query structures using a tuple in case we run out of names :), (credits to @michiyasunaga). For example, 1p queries: (e, (r,)) and 2i queries: ((e, (r,)),(e, (r,))). Check the code for more details.

**Examples**

Please refer to the `examples.sh` for the scripts of all 3 models on all 3 datasets.

**Citations**

If you use this repo, please cite the following paper.

```

@inproceedings{

 ren2020beta,

 title={Beta Embeddings for Multi-Hop Logical Reasoning in Knowledge Graphs},

 author={Hongyu Ren and Jure Leskovec},

 booktitle={Neural Information Processing Systems},

 year={2020}

}

```