Open Access Highly Accessed Open Badges Research article

A pipeline to extract drug-adverse event pairs from multiple data sources

SriJyothsna Yeleswarapu, Aditya Rao*, Thomas Joseph, Vangala Govindakrishnan Saipradeep and Rajgopal Srinivasan

Author Affiliations

TCS Innovation Labs, Tata Consultancy Services Ltd, Deccan Park, 1, Software Units Layout, Madhapur, Hyderabad 500081, Andhra Pradesh, India

For all author emails, please log on.

BMC Medical Informatics and Decision Making 2014, 14:13  doi:10.1186/1472-6947-14-13

Published: 24 February 2014



Pharmacovigilance aims to uncover and understand harmful side-effects of drugs, termed adverse events (AEs). Although the current process of pharmacovigilance is very systematic, the increasing amount of information available in specialized health-related websites as well as the exponential growth in medical literature presents a unique opportunity to supplement traditional adverse event gathering mechanisms with new-age ones.


We present a semi-automated pipeline to extract associations between drugs and side effects from traditional structured adverse event databases, enhanced by potential drug-adverse event pairs mined from user-comments from health-related websites and MEDLINE abstracts. The pipeline was tested using a set of 12 drugs representative of two previous studies of adverse event extraction from health-related websites and MEDLINE abstracts.


Testing the pipeline shows that mining non-traditional sources helps substantiate the adverse event databases. The non-traditional sources not only contain the known AEs, but also suggest some unreported AEs for drugs which can then be analyzed further.


A semi-automated pipeline to extract the AE pairs from adverse event databases as well as potential AE pairs from non-traditional sources such as text from MEDLINE abstracts and user-comments from health-related websites is presented.

Pharmacovigilance; NLP; Text mining; Social media; Adverse event; Biomedical literature; Unstructured text; BCPNN