Pharmacovigilance from social media: mining adverse drug reaction mentions using sequence labeling with word embedding cluster features

Abstract

Objective: Social media is becoming increasingly popular as a platform for sharing personal health-related information. This information can be utilized for public health monitoring tasks, particularly for pharmacovigilance, via the use of natural language processing (NLP) techniques. However, the language in social media is highly informal, and user-expressed medical concepts are often nontechnical, descriptive, and challenging to extract. There has been limited progress in addressing these challenges, and thus far, advanced machine learning-based NLP techniques have been underutilized. Our objective is to design a machine learning-based approach to extract mentions of adverse drug reactions (ADRs) from highly informal text in social media.

Methods: We introduce ADRMine, a machine learning-based concept extraction system that uses conditional random fields (CRFs). ADRMine utilizes a variety of features, including a novel feature for modeling words' semantic similarities. The similarities are modeled by clustering words based on unsupervised, pretrained word representation vectors (embeddings) generated from unlabeled user posts in social media using a deep learning technique.

Results: ADRMine outperforms several strong baseline systems in the ADR extraction task by achieving an F-measure of 0.82. Feature analysis demonstrates that the proposed word cluster features significantly improve extraction performance.

Conclusion: It is possible to extract complex medical concepts, with relatively high performance, from informal, user-generated content. Our approach is particularly scalable, suitable for social media mining, as it relies on large volumes of unlabeled data, thus diminishing the need for large, annotated training data sets.

Keywords: ADR; adverse drug reaction; deep learning word embeddings; machine learning; natural language processing; pharmacovigilance; social media mining.

Figure 1

Examples of user-posted drug reviews in Twitter (a) and DailyStrength (b).

Figure 2

Calculated features representing three CRF classification instances.

Figure 3

The impact of embedding clusters on precision, recall (a), and F-measure (b), when training the CRF on variable training set sizes and testing on the same test set.

Figure 4

Examples of successfully extracted concepts using ADRMine.

Figure 5

Examples of concepts that could only be extracted after adding the embedding cluster features to ADRMine. These concepts are starred and other extracted concepts are highlighted.

Figure 6

Analysis of false positive and false negatives produced by the ADR extraction approach.