TDGNN：Continuous-Time Link Prediction via Temporal Dependent Graph Neural Network

This is our Pytorch implementation for the paper:

Liang Qu, Huaisheng Zhu, Qiqi Duan, and Yuhui Shi. 2020. Continuous-Time Link Prediction via Temporal Dependent Graph Neural Network. In Proceedings of The Web Conference 2020 (WWW '20). Association for Computing Machinery, New York, NY, USA, 3026–3032. DOI:https://doi.org/10.1145/3366423.3380073

Introduction

Temporal Dependent Graph Neural Network (TDGNN), a simple yet effective dynamic network representation learning framework which incorporates the network temporal information into GNNs. TDGNN introduces a novel Temporal Aggregator (TDAgg) to aggregate the neighbor nodes’ features and edges’ temporal information to obtain the target node representations.

Citation

If you want to use our codes in your research, please cite:

@inproceedings{10.1145/3366423.3380073,
author = {Qu, Liang and Zhu, Huaisheng and Duan, Qiqi and Shi, Yuhui},
title = {Continuous-Time Link Prediction via Temporal Dependent Graph Neural Network},
year = {2020},
isbn = {9781450370233},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3366423.3380073},
doi = {10.1145/3366423.3380073},
booktitle = {Proceedings of The Web Conference 2020},
pages = {3026–3032},
numpages = {7},
location = {Taipei, Taiwan},
series = {WWW '20}
}

Usage

python3 model.py -input_node ../contact/feature_random_contact.txt -input_edge_train ../contact/edge_train_contact -input_edge_test ../contact/edge_train_contact -output_file result -aggregate_function origin -hidden_dimension 128