Ensemble-Based Medical Relation Classification

Jennifer D'Souza and Vincent Ng.
Proceedings of the 25th International Conference on Computational Linguistics, pp. 1682-1693, 2014.

Click here for the PostScript or PDF version. The talk slides are available here.

Abstract

Despite the successes of distant supervision approaches to relation extraction in the news domain, the lack of a comprehensive ontology of medical relations makes it difficult to apply such approaches to relation classification in the medical domain. In light of this difficulty, we propose an ensemble approach to this task where we exploit human-supplied knowledge to guide the design of members of the ensemble. Results on the 2010 i2b2/VA Challenge corpus show that our ensemble approach yields a 19.8% relative error reduction over a state-of-the-art baseline.

Ruleset

The hand-crafted rules described in Section 4.3 of the paper are available from this page.

BibTeX entry

@InProceedings{DSouza+Ng:14b,
  author = {Jennifer D'Souza and Vincent Ng},
  title = {Ensemble-Based Medical Relation Classification},
  booktitle = {Proceedings of the 25th International Conference on Computational Linguistics},
  pages = {1682--1693}, 
  year = 2014}