CovEx: A User-Centric Recommender System for COVID-19 Scientific Literature

Contact Email
Abstract

CovEx system has been deployed online and demonstrated to several target users. The early results indicate that the success of the system to a considerable extent depends on the quality of key phrase extraction. Moreover, the nature of exploratory search calls for special extraction approaches. While we used a relatively powerful approach, it was trained to model gold standard annotation of individual documents in GENIA dataset. We believe, however, that key phrase extraction has to consider the collection as a whole increasing user chances to discover key phrases that could lead to other papers. We are converting CovEx to support clinicians' exploration of Ovarian cancer related literature to further examination of usages of CovEx.

Authors
Young Ji
Lee
Assistant Professor
Behnam
Rahdari
PhD Sudent
Peter
Brusilovsky
Professor
Khushboo
Thaker
PhD Student