Clinical NER and Relation Extraction using Bi-Char-LSTMs and Random Forest Classifiers

[edit]

Arjun Magge, Matthew Scotch, Graciela Gonzalez-Hernandez ;
Proceedings of the 1st International Workshop on Medication and Adverse Drug Event Detection, PMLR 90:25-30, 2018.

Abstract

Identifying named entities from electronic health record notes and extracting relations between the entities is a crucial task for many applications in clinical and public health informatics. In this work, we present an natural language processing pipeline consisting of a named entity recognizer for identifying 9 medical named entities in clinical notes and a random forests classifier for extracting 7 types of relations between the entities.

Related Material