Expert Finding in Legal Community Question Answering

Arian Askari, Suzan Verberne, and Gabriella Pasi. Expert Finding in Legal Community Question Answering. ECIR 2022 (short).

If you use this work, please cite as:

@inproceedings{askari2022expert,
  title={Expert Finding in Legal Community Question Answering},
  author={Askari, Arian and Verberne, Suzan and Pasi, Gabriella},
  booktitle={European Conference on Information Retrieval},
  pages={22--30},
  year={2022},
  organization={Springer}
}

Dataset

Queries

All queries are are available in /data/queries_bankruptcy.csv

Posts

The link of all posts that have been used for this reaserch (in Bankruptcy category) are available in /data/question_links_bankruptcy.json

Labels

The labels are provided in qrel format (queyr_id itteration user_id label) in /data/labels.qrel. Itteration is always zero and not used. Query id refers to id of query in queries_bankruptcy.csv file, user id refers to the lawyer id, label is zero for non-expert and one for expert users.

Lawyers' webpage on Avvo

The lawyers' webpage addresses are available in lawyerid_to_lawyerurl.json in {"user id": "lawyer url"} format. Therefore, the lawyer ids (user ids) in labels.csv can be mapped to their webpage on Avvo by this file.

Baselines

The implementation of candidate level and document level baselines (Model 1 LM, and Model 2 LM) using Elasticsearch are available at baselines/candidate_level_lm_lablog.py and baselines/document_level_lm_balog.py.

P.S: All pages were stored anonymously during this research with regard to the users' privacy.