This repository contains the accompanying code for the paper, where we present a query-based biomedical information retrieval task across two vastly different genres – newswire and research literature – where the goal is to find the research publication that supports the primary claim madein a health-related news article. For this task,we present a new dataset of 5,034 claims fromnews paired with research abstracts.
The dataset for the Cross Genre IR Task along with other supporting files can be accessed here.
Our approach consists of two steps:
- Selecting the most relevant candidates from a collection of 222k research abstracts, and
- Re-rankingthis the candidates list.
If you use this work please cite our research paper:
@inproceedings{Zuo2020QueryingAG,
title={Querying across Genres to Retrieve Research That Supports Medical Claims Made in News},
author={Chaoyuan Zuo and Narayan Acharya and Ritwik Banerjee},
booktitle={EMNLP},
year={2020}
}