Skip to main content
Fig. 1 | BMC Bioinformatics

Fig. 1

From: Leveraging graph topology and semantic context for pharmacovigilance through twitter-streams

Fig. 1

Method Overview. Two DEGs are constructed from two datasets, SIDER (left) and Twitter (Right) 1. DEGSider is a graph built from the drug and effect relationships recorded in the SIDER database, forming a comprehensive DEG containing the Adverse Drug Relationships known to SIDER. 2. DEGTwitter is constructed by mining Twitter posts for mentions of drugs and effects. 3. The live Twitter-stream is filtered for tweets that reference drugs and relate an individual user’s experience. 4. From the tweets gathered from the Twitter-stream, user’s histories are queried. 5. Drug and effect mentions are extracted with MetaMap. 6. An edge is drawn between a drug and an effect if they are both referenced in a user’s history within a fixed window of time. These edges are labeled as adverse or psuedo-non-adverse based on their presence or absence in SIDER

Back to article page