CEM System for Chemical Entity Mention Recognition in Patents

In this paper, a machine learning-based system was proposed for the challenge task of chemical entity mention recognition in patents (CEMP) in BioCreative V. The CEMP task was recognized as a sequence labeling problem and conditional random fields (CRF) were employed for it. Evaluation on the CEMP challenge corpus showed that our system (team 293) achieved a micro F-measure of 87.03%.