A step-by-step Keras implementation of PACE (Preference And Context Embedding) described in our KDD 2017 paper.
Switch branches/tags
Nothing to show
Clone or download
Latest commit 7ed7c3c Dec 5, 2017
Permalink
Failed to load latest commit information.
README.md Create README.md Jun 13, 2017
dataset.py Upload git files Jun 9, 2017
download_data.sh Update download_data.sh Dec 5, 2017
train.ipynb Upload git files Jun 9, 2017
utils.py Upload git files Jun 9, 2017

README.md

PACE

A step-by-step Keras implementation of PACE (Preference And Context Embedding) described in our KDD 2017 paper. To run the code, you need to have Python 3 and iPython Notebook installed.

Please cite the following work.

Carl Yang, Lanxiao Bai, Chao Zhang, Quan Yuan and Jiawei Han. 2017. Bridging Collaborative Filtering and Semi-Supervised Learning: A Neural Approach for POI Recommendation. In Proceedings of KDD ?17, Halifax, NS, Canada, August 13-17, 2017, 10 pages.

Usage:

  • Use bash download_data.sh to download the Gowalla data or visit Yelp to download the Yelp data.
  • Run python3 dataset.py for data preprocessing (slight modifications needed to match specific data formats).
  • Start iPython Notebook Server ipython3 notebook
  • Sequentially run cells in train.ipynb

If you are using remote machine, you can:

  • Start iPython Notebook Server on remote machine: ipython notebook --no-browser --port=8889
  • Redirect ssh connection to localhost ssh -N -f -L localhost:8880:localhost:8889 <user>@<host>
  • Open browser and go to <user>@<host>:8880