This is the pytorch implementation of paper "Learning Robust Travel Preferences via Check-in Masking for Next POI Recommendation"
pip install -r requirements.txt
torch==2.0.1
numpy==1.24.3
pandas==2.0.2
Pillow==9.4.0
python-dateutil==2.8.2
pytz==2023.3
six==1.16.0
torchvision==0.15.2
typing_extensions==4.5.0
-
Unzip
raw_data/raw_data.ziptoraw_data/. The three files are PHO, NYC and SIN check-in data. -
Run
data_preprocessor.pyto construct input data.- Or unzip
processed_data/processed_data.ziptoprocessed_data/.
- Or unzip
-
Train and evaluate the model using python
main.py. -
The training and evaluation results will be stored in
resultfolder.
python main.py --name NYC_auto0.4
--run-times 1
--device cuda:0
--dataset NYC
--mask-strategy 0 --mask-proportion 0.4
--area-proportion 0.2
--embed-size 60
--transformer-layers 1 --transformer-heads 1
--dropout 0.2
--epochs 40
--lr 1e-5
@article{RTP-CM2024,
title = {Learning robust travel preferences via check-in masking for next {POI} recommendation},
issn = {0957-4174},
doi = {10.1016/j.eswa.2024.126106},
pages = {126106},
journaltitle = {Expert Systems with Applications},
shortjournal = {Expert Systems with Applications},
author = {Duan, Chenghua and Wen, Junhao and Zhou, Wei and Zeng, Jun and Zhang, Yihao},
date = {2024-12-18},
keywords = {Masking strategy, Noise reduction, {POI} recommendation}
}