Python >= 3.6

pip3 install -r requirements.txt
python3 -m spacy download en

1. setup elasticsearch service, refer to link

2. setting value ES_BASE_URL in constants.py with your configured elastic search endpoint.

1. unzip file and put all files under data/ folder, rename test.csv to test_release.csv
2. execute bash scripts/prepare_data.sh in project root folder to build the data for next step

execute bash scripts/run_retrieval.sh in project root folder

the above script includes three main parts

1. execute elasticsearch to retrieval candidate papers

2. prepare rerank data from elastic search result (baseline result)

3. execute the rerank by BERT

1. recall phase

   noun chunk extraction + textrank keyword extraction + BM25 based search (elasticsearch)

2. rerank phase

   Bert based rerank (SciBert from AllenAI)

The model required to be trained in this project just the Bert based reranking model