Extracting Consumer Health Expressions of Drug Safety from Web Forum

Consumers often have difficulties expressing and understanding medical terminologies due to gaps in their domain knowledge and those of the health care professionals. This language gap is a barrier to effective health information seeking, and ultimately, informed decision making. However, despite the recent research on mismatches between consumer and professional languages, there have been limited studies tackles on how to effectively discover consumer health-related expressions through mining of social media data. In this research, we propose an automatic key-phrase extraction approach to identify consumer health expressions with regard to adverse drug reaction (ADRs) in social media. These identified expressions can help to extend current Consumer Health Vocabularies (CHV) and to enhance the performance of ADRs signal detection for pharmacovigilance systems. The proposed method can be applied to other problem domains that require automatic key-phrase extraction when there is a mismatch between the languages used by the layperson and the professionals.