Paper: Towards Automated Semantic Analysis on Biomedical Research Articles

ACL ID I08-2124
Title Towards Automated Semantic Analysis on Biomedical Research Articles
Venue International Joint Conference on Natural Language Processing
Session Main Conference
Year 2008
Authors

In this paper, we present an empirical study on adapting Conditional Random Fields (CRF) models to conduct semantic analysis on biomedical articles using ac- tive learning. We explore uncertainty- based active learning with the CRF model to dynamically select the most informa- tive training examples. This abridges the power of the supervised methods and ex- pensive human annotation cost.