COVID-19: Comparative Analysis of Methods for Identifying Articles Related to Therapeutics and Vaccines without Using Labeled Data

2021/01/05 by Mihir Parmar, Parmar, Mihir, Ashwin Karthik Ambalavanan +9
Medicine · Pharmacology, Toxicology and Pharmaceutics · Social Sciences · #Data-Driven Disease Surveillance #FOS: Computer and information sciences #Information Retrieval (cs.IR) #Machine Learning (cs.LG) #Pharmacovigilance and Adverse Drug Reactions #Vaccine Coverage and Hesitancy

paper · pdf · doi:10.48550/arxiv.2101.02017

openalex publication_date 2021/01/05 · openalex created_date 2021/01/18 · openalex updated_date 2026/07/28

Abstract

Here we proposed an approach to analyze text classification methods based on the presence or absence of task-specific terms (and their synonyms) in the text. We applied this approach to study six different transfer-learning and unsupervised methods for screening articles relevant to COVID-19 vaccines and therapeutics. The analysis revealed that while a BERT model trained on search-engine results generally performed well, it miss-classified relevant abstracts that did not contain task-specific terms. We used this insight to create a more effective unsupervised ensemble.

Related

• Document Classification for COVID-19 Literature
• Getting Insights from a Large Corpus of Scientific Papers on Specialisted Comprehensive Topics -- the Case of COVID-19
• Non-Pharmaceutical Intervention Discovery with Topic Modeling
• Automated Multilabel Mpox Research Classification with Explainable Transformer Models
• Actively evaluating and learning the distinctions that matter: Vaccine safety signal detection from emergency triage notes