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.
@article{arxiv.2101.02017,
title = {COVID-19: Comparative Analysis of Methods for Identifying Articles Related to Therapeutics and Vaccines without Using Labeled Data},
author = {Mihir Parmar and Ashwin Karthik Ambalavanan and Hong Guan and Rishab Banerjee and Jitesh Pabla and Murthy Devarakonda},
journal= {arXiv preprint arXiv:2101.02017},
year = {2021}
}