Skip to content

Latest commit

 

History

History
41 lines (23 loc) · 1.15 KB

README.md

File metadata and controls

41 lines (23 loc) · 1.15 KB

ReviewSum

Codes and datasets for our paper: Neural Review Summarization Leveraging User and Product Information (CIKM19)

Requirements

  • python >= 3.5
  • pytorch >= 1.1.0
  • sumeval
  • tqdm

Datasets

Download the datasets from Google Drive or Baidu Pan. Put the unziped /data directory into the project root directory.

Models

The repository contains three baseline models: seq2seq, seq2seqAttn, pgn, and our proposed four models: AttrEnc, AttrDec, AttrEncDec, and MemAttr, as mentioned in our paper. The command to run each model is the same. Take our novel model MemAttr as an example:

Train a MemAttr model:

cd code/memAttr
python train.py 

Test the trained model:

python train.py -test -load_model <the_checkpoint_you_want_to_test>

The train and test parameters can be found in the source code train.py.

Citation

If you use our codes or datasets in your research, please kindly cite our paper: Neural Review Summarization Leveraging User and Product Information (CIKM19)