This project builds upon KP4SR, which serves as our baseline model.

1. Clone KP4SR into this directory.
2. Replace the corresponding files in the scripts and src directories with those provided here.

• Masked Personalized Prompt
• Venue Knowledge Prompt
• User Knowledge Prompt

1. Convert the original data and add additional information.
   • See the README in the preprocess directory.
2. Prepare the data required by KP4SR.
   • See the README in the data directory for experimental setup.

Once the data are prepared, run:

bash run_poi.sh

If you find this work useful, please cite our paper:

@inproceedings{Tachioka2025_RecTour,
  author    = {Yuuki Tachioka},
  title     = {{KP4POI}: Efficient POI Recommendation on Large-Scale Datasets via Knowledge Prompting of Venues and Users},
  booktitle = {Proceedings of the ACM RecSys Workshop on Recommenders in Tourism @ 19th ACM Conference on Recommender Systems (RecSys 2025)},
  location  = {Prague, Czech Republic},
  month     = {September},
  year      = {2025},
}