Embedding Uncertain Knowledge Graphs

This repository includes the code of UKGE and data used in the experiments.

Install

Make sure your local environment has the following installed:

Python3
tensorflow >= 1.5.0
scikit-learn

Install the dependents using:

pip install -r requirements.txt

Run the experiments

To run the experiments, use:

python ./run/run.py

or

python ./run/run.py --data ppi5k --model rect --batch_size 1024 --dim 128 --epoch 100 --reg_scale 5e-4

You can use --model logi to switch to the UKGE(logi) model.

Data is available at: https://drive.google.com/file/d/1UJQ8hnqPGv1O9pYglfNF5lY_sgDQkleS/view?usp=sharing

Reference

Please refer to our paper. Xuelu Chen, Muhao Chen, Weijia Shi, Yizhou Sun, Carlo Zaniolo. Embedding Uncertain Knowledge Graphs. In Proceedings of the Thirty-Third AAAI Conference on Artificial Intelligence (AAAI), 2019

@inproceedings{chen2019ucgraph,
    title={Embedding Uncertain Knowledge Graphs},
    author={Chen, Xuelu and Chen, Muhao and Shi, Weijia and Sun, Yizhou and Zaniolo, Carlo},
    booktitle={Proceedings of the Thirty-Third AAAI Conference on Artificial Intelligence (AAAI)},
    year={2019}
}