Biocomputing and Media Research Lab

Image: Figures pertaining to research

Leveraging Graph Topology and Semantic Context for Pharmacovigilance in Twitter Streams

Description

Adverse drug events (ADEs) constitute one of the leading causes of post-therapeutic death and their identification constitutes an important challenge of modern precision medicine. Unfortunately, the onset and effects of ADEs are often underreported complicating timely intervention. At over 500 million posts per day, Twitter is a commonly used social media platform. The ubiquity of day-to-day personal information exchange on Twitter makes it a promising target for data mining for ADE identification and intervention. Three technical challenges are central to this problem: (1) identification of salient medical keywords in (noisy) tweets, (2) mapping drug-effect relationships, and (3) classification of such relationships as adverse or non-adverse. We first use dictionary-based and algorithmic information extraction to identify occurrences of medically-relevant concepts in tweets. Next, a drug-effect graph (DEG) is constructed by mining users’ tweet history. In the DEG, drugs and symptoms are connected with edges weighted by temporal distance and frequency. From this graph, edges are classified as either adverse or non-adverse with a classifier trained using both graph-theoretic and semantic features such as sentiment polarity of the source text.

Publications

  • R. Eshleman, and R. Singh, "Leveraging Graph Topology and Semantic Context for Pharmacovigilance in Twitter Streams", Midsouth Computational Biology and Bioinformatics Society Conference (MCBIOS) 2016, peer reviewed abstract, 2016
SF State Home