Modeling Drug-Disease Relations with Linguistic and Knowledge Graph Constraints
This work addresses the problem of extracting drug-disease relations from complex texts for healthcare and pharmaceutical applications, representing an incremental improvement through hybrid modeling.
The researchers tackled the challenge of analyzing complex FDA drug labels by integrating them with health knowledge graphs, showing that Probabilistic Soft Logic models combining linguistic and graph constraints outperform text-only and relation-only methods, with the clinical narratives graph achieving exceptional results using minimal manual effort.
FDA drug labels are rich sources of information about drugs and drug-disease relations, but their complexity makes them challenging texts to analyze in isolation. To overcome this, we situate these labels in two health knowledge graphs: one built from precise structured information about drugs and diseases, and another built entirely from a database of clinical narrative texts using simple heuristic methods. We show that Probabilistic Soft Logic models defined over these graphs are superior to text-only and relation-only variants, and that the clinical narratives graph delivers exceptional results with little manual effort. Finally, we release a new dataset of drug labels with annotations for five distinct drug-disease relations.