For the submission of BIOSTAT 823 Final Project
This pipeline is developed based on Stanford CoreNLP, Stanza, and Unified Medical Language System (UMLS). Given textual input, this pipeline extracts triple relations that contain medical entities, and search for UMLS concept unique identifiers (CUI) that are relevant to the entities. Based on the extracted triples and CUIs, this pipeline can compute the distance matrix based on each entity's relevant CUIs, compute the transitive closure and distance matrices with a given relation / predicate.
Instructors and teaching assistant of BIOSTAT 823