Paper: Learning to Detect Hedges and their Scope Using CRF

ACL ID W10-3014
Title Learning to Detect Hedges and their Scope Using CRF
Venue International Conference on Computational Natural Language Learning
Session Main Conference
Year 2010
Authors

Detecting speculative assertions is essential to distinguish the facts from uncertain information for biomedical text. This paper describes a system to detect hedge cues and their scope using CRF model. HCDic feature is presented to improve the system perfor- mance of detecting hedge cues on BioScope corpus. The feature can make use of cross- domain resources.