Extraction of Medication Names from Twitter Using Augmentation and an Ensemble of Language Models

TL;DR

This paper presents a method that combines data augmentation and an ensemble of pre-trained language models to improve medication name extraction from Twitter, achieving state-of-the-art results in a challenging benchmark.

Contribution

The study introduces a novel combination of data augmentation and ensemble modeling for medication extraction from social media, surpassing previous methods.

Findings

Outperformed the Kusuri algorithm in the BioCreative VII challenge

Achieved high overlapping F1 score in medication name extraction

Demonstrated effectiveness of data augmentation and ensemble models

Abstract

The BioCreative VII Track 3 challenge focused on the identification of medication names in Twitter user timelines. For our submission to this challenge, we expanded the available training data by using several data augmentation techniques. The augmented data was then used to fine-tune an ensemble of language models that had been pre-trained on general-domain Twitter content. The proposed approach outperformed the prior state-of-the-art algorithm Kusuri and ranked high in the competition for our selected objective function, overlapping F1 score.