accepted by ACL2020: Hiring Now: A Skill-Aware Multi-Attention Model for Job Posting Generation
the environment:
-
python==3.7.6
-
torch==1.2.0
-
cudatoolkit==10.0.130
-
the trained model and data: [url]
download the data and model.
-
unzip data_model.zip mv *pt SAMA/dataset mv pretrained_w2v SAMA/dataset mv trainedmodel SAMA/trained_model
-
train or just reproduce the results
python train.py --file_dir SAMA/dataset # for training python reproduce.py --file_dir SAMA/dataset # just reproduce
- src -- the text of job description
- srcid -- the word id of job description
- tgt -- the text of skill requirement
- tgtid -- the word id of skill requirement
- skillnet -- the recommed skill text of skill knowledge graph
- skillnetid -- the recommed skill word id of skill knowledge graph
- skilltgt -- the extracted skill text of skill requirement
- skilltgtid -- the extracted skill word id of skill requirement