Finding potentially unsafe nutritional supplements from user reviews with topic modeling

Ryan Sullivan, Abeed Sarker, Karen O'connor, Amanda Goodin, Mark Karlsrud, Graciela Gonzalez

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contributionpeer-review

10 Citations (Scopus)
7 Downloads (Pure)

Abstract

Although dietary supplements are widely used and generally are considered safe, some supplements have been identified as causative agents for adverse reactions, some of which may even be fatal. The Food and Drug Administration (FDA) is responsible for monitoring supplements and ensuring that supplements are safe. However, current surveillance protocols are not always effective. Leveraging user-generated textual data, in the form of Amazon.com reviews for nutritional supplements, we use natural language processing techniques to develop a system for the monitoring of dietary sup- plements. We use topic modeling techniques, specifically a variation of Latent Dirichlet Allocation (LDA), and background knowledge in the form of an adverse reaction dictionary to score products based on their potential danger to the public. Our approach generates topics that semantically cap- ture adverse reactions from a document set consisting of reviews posted by users of specific products, and based on these topics, we propose a scoring mechanism to categorize products as “high potential danger”, “average potential danger” and “low potential danger.” We evaluate our system by com- paring the system categorization with human annotators, and we find that the our system agrees with the annotators 69.4% of the time. With these results, we demonstrate that our methods show promise and that our system represents a proof of concept as a viable low-cost, active approach for dietary supplement monitoring.

Original languageEnglish
Title of host publicationPacific Symposium on Biocomputing 2016
EditorsRuss B. Altman, A. Keith Dunker, Lawrence Hunter, Marylyn D. Ritchie, Tiffany Murray, Teri E. Klein
Place of PublicationSingapore
PublisherWorld Scientific Publishing
Pages528-539
Number of pages12
ISBN (Electronic)9789814749411, 9789814749411
ISBN (Print)9789814749404
DOIs
Publication statusPublished - 2016
Externally publishedYes
Event21st Pacific Symposium on Biocomputing, PSB 2016 - Kohala Coast, United States
Duration: 4 Jan 20168 Jan 2016

Conference

Conference21st Pacific Symposium on Biocomputing, PSB 2016
CountryUnited States
CityKohala Coast
Period4/01/168/01/16

Keywords

  • Dietary supplements
  • Latent dirichlet allocation
  • Natural language processing
  • Pharmacovigilance
  • Public health surveillance
  • Social media mining

Fingerprint Dive into the research topics of 'Finding potentially unsafe nutritional supplements from user reviews with topic modeling'. Together they form a unique fingerprint.

Cite this