Co-training for Extraction of Adverse Drug Reaction Mentions from Tweets

TL;DR

This paper introduces a semi-supervised co-training approach to extract adverse drug reaction mentions from tweets, leveraging unlabeled data to improve performance over existing RNN-based methods.

Contribution

It presents a novel co-training method that effectively utilizes unlabeled social media data to enhance ADR mention extraction accuracy.

Findings

Outperforms state-of-the-art methods by 5% F1 score

Uses 0.1 million tweets for training and evaluation

Demonstrates the effectiveness of semi-supervised learning in ADR extraction

Abstract

Adverse drug reactions (ADRs) are one of the leading causes of mortality in health care. Current ADR surveillance systems are often associated with a substantial time lag before such events are officially published. On the other hand, online social media such as Twitter contain information about ADR events in real-time, much before any official reporting. Current state-of-the-art methods in ADR mention extraction use Recurrent Neural Networks (RNN), which typically need large labeled corpora. Towards this end, we propose a semi-supervised method based on co-training which can exploit a large pool of unlabeled tweets to augment the limited supervised training data, and as a result enhance the performance. Experiments with 0.1M tweets show that the proposed approach outperforms the state-of-the-art methods for the ADR mention extraction task by 5% in terms of F1 score.