Element-guided Temporal Graph Representation Learning for Temporal Sets Prediction

The description of "Element-guided Temporal Graph Representation Learning for Temporal Sets Prediction" at WWW 2022 is available here.

Original data:

The original data could be downloaded from here. You can download the data and then put the data files in the ./original_data folder.

To run the code:

1. run ./preprocess_data/preprocess_data_{dataset_name}.py to preprocess the original data, where dataset_name could be DC, TaoBao, JingDong and TMS. We also provide the preprocessed datasets at here, which should be put in the ./dataset folder.

2. run ./train/train_ETGNN.py to train the model on different datasets using the configuration in ./utils/config.json.

3. run ./evaluate/evaluate_ETGNN.py to evaluate the model. Please make sure the config in evaluate_ETGNN.py keeps identical to that in the model training process.

Environments:

• PyTorch 1.8.1
• tqdm
• numpy
• pandas

Hyperparameter settings:

Hyperparameters can be found in ./utils/config.json file, and you can adjust them when training the model on different datasets.

| Hyperparameters | DC | TaoBao | JingDong | TMS | | ------- | ------- | ------- | ------- | ------- | | learning rate | 0.001 | 0.001 | 0.001 | 0.001 | | embedding dimension | 64 | 32 | 64 | 64 | | embedding dropout | 0.2 | 0.0 | 0.2 | 0.3 | | temporal attention dropout | 0.5 | 0.5 | 0.5 | 0.5 | | number of hops | 3 | 3 | 3 | 2 | | temporal information importance | 0.3 | 0.05 | 0.01 | 1.0 |

Citation

Please consider citing our paper when using the codes or datasets.

@inproceedings{DBLP:conf/www/YuWS0L22,
  author    = {Le Yu and
               Guanghui Wu and
               Leilei Sun and
               Bowen Du and
               Weifeng Lv},
  title     = {Element-guided Temporal Graph Representation Learning for Temporal
               Sets Prediction},
  booktitle = {{WWW} '22: The {ACM} Web Conference 2022, Virtual Event, Lyon, France,
               April 25 - 29, 2022},
  pages     = {1902--1913},
  publisher = {{ACM}},
  year      = {2022}
}