Re-ranking Biomedical Literature for Precision Medicine with Pre-trained Neural Models

被引：0

作者：

Jiazhao Li ^{[1
,2
]}

Murali, Adharsh ^{[2
]}

Qiaozhu Mei ^{[1
]}

Vydiswaran, V. G. Vinod ^{[1
,2
]}

机构：

[1] Univ Michigan, Sch Informat, Ann Arbor, MI USA

[2] Univ Michigan, Med Sch, Dept Learning Hlth Sci, Ann Arbor, MI USA

来源：

2020 8TH IEEE INTERNATIONAL CONFERENCE ON HEALTHCARE INFORMATICS (ICHI 2020) | 2020年

关键词：

biomedical retrieval; pretrained models; BERT;

D O I：

10.1109/ICHI48887.2020.9374401

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We propose a biomedical literature retrieval approach that incorporates a domain-specific BERT model as an auxiliary re-ranker. Experiments on TREC Precision Medicine dataset show its effectiveness in improving retrieval performance by 6.2% in inferred NDCG and 6.8% in R-precision over the best-published results. The contribution of this study is to provide evidence of incorporating BERT in a biomedical literature retrieval system, which serves the overall goal to improve the information retrieval for precision medicine.

引用

页码：511 / 513

页数：3

共 13 条

[1] Reciprocal Rank Fusion outperforms Condorcet and Individual Rank Learning Methods
Cormack, Gordon V.
Clarke, Charles L. A.
Buettcher, Stefan
[J]. PROCEEDINGS 32ND ANNUAL INTERNATIONAL ACM SIGIR CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION RETRIEVAL, 2009, : 758 - 759
[2] Devlin J, 2019, Arxiv, DOI arXiv:1810.04805
[3] Goodwin T., 2017, Proceedings of TREC-PM, P1
[4] Lee Jinhyuk, 2019, arXiv
[5] Liu J., 2018, P 27 TEXT RETRIEVAL
[6] Nogueira R, 2020, Arxiv, DOI arXiv:1901.04085
[7] Oleynik Michel., 2018, TREC
[8] Roberts Kirk, 2018, NIST SPECIAL PUBLICA, V500-331
[9] Denoising Clinical Notes for Medical Literature Retrieval with Convolutional Neural Model
Soldaini, Luca
Yates, Andrew
Goharian, Nazli
[J]. CIKM'17: PROCEEDINGS OF THE 2017 ACM CONFERENCE ON INFORMATION AND KNOWLEDGE MANAGEMENT, 2017, : 2307 - 2310
[10] Wu YH, 2016, Arxiv, DOI arXiv:1609.08144

← 1 2 →