Extracting drug indications and adverse drug reactions from Spanish health social media

Abstract

In this paper, we present preliminary results obtained using a system based on co-occurrence of drug-effect pairs as a first step in the study of detecting adverse drug reactions and drug indications from social media texts. To the best of our knowledge, this is the first work that extracts this kind of relationships from user messages that were collected from an online Spanish health-forum. In addition, we also describe the automatic construction of the first Spanish database for drug indications and adverse drug reactions.