A Sui Generis QA Approach using RoBERTa for Adverse Drug Event Identification

TL;DR

This paper introduces a novel question answering approach using RoBERTa, tailored for adverse drug event identification in biomedical texts, achieving significant performance improvements over previous entity-relation extraction methods.

Contribution

The paper presents a domain-adapted RoBERTa-based QA framework that surpasses prior models in adverse drug event detection, addressing limitations of traditional Bi-LSTM approaches.

Findings

Model outperforms previous methods by 9.53% F1-Score

Utilizes RoBERTa's robustness and dynamic attention for better feature extraction

Effective domain adaptation enhances biomedical text analysis

Abstract

Extraction of adverse drug events from biomedical literature and other textual data is an important component to monitor drug-safety and this has attracted attention of many researchers in healthcare. Existing works are more pivoted around entity-relation extraction using bidirectional long short term memory networks (Bi-LSTM) which does not attain the best feature representations. In this paper, we introduce a question answering framework that exploits the robustness, masking and dynamic attention capabilities of RoBERTa by a technique of domain adaptation and attempt to overcome the aforementioned limitations. Our model outperforms the prior work by 9.53% F1-Score.