DetGP

This repository contains the source code to reproduce the results of DetGP: Dynamic Embedding on Textual Networks via a Gaussian Process.

Here is the link to our paper:

• Dynamic Embedding on Textual Networks via a Gaussian Process (AAAI 2020)

This project is maintained by Pengyu Cheng. Feel free to contact pengyu.cheng@duke.edu for any relevant issues.

Dependencies:

This code is written in python. The dependencies are:

• Python 2.7
• Tensorflow>=1.13 (1.13.1 is recommended)
• networkx

Download Data

Download the folder ./datasets/ from the CANE repository, which includes Cora and HepTh datasets.

Run the code

To reproduce the link prediction results of DetGP, run the command:

python train_link_predict.py

To reproduce the node classification results of DetGP, run the command:

python train_node_classify.py

The parameters for the experiments can be set in config.py, e.g., the choice of datasets, the train-test split ratios.

Citation

Please cite our AAAI 2020 paper if you found the code useful.

@misc{cheng2019dynamic,
    title={Dynamic Embedding on Textual Networks via a Gaussian Process},
    author={Pengyu Cheng and Yitong Li and Xinyuan Zhang and Liqun Cheng and David Carlson and Lawrence Carin},
    year={2019},
    eprint={1910.02187},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}