Improving the performance of bio-entity name recognition in biomedical literature via the contextual cues

摘要

To extract biomedical information about bio-entities from the huge amount of biomedical literature, the first key step is recognizing the bio-entity names in literature, which remains a challenging task due to the irregularities and ambiguities in bio-entities nomenclature. Currently the performance of the most popular methods, machine learning techniques, still have much room for improvement. This paper presents a Conditional Random Fields-based bio-entity name recognition approach and studies the methods of improving the performance by exploiting contextual cues and other post-processing methods. Experiment results show that these methods improve Conditional Random Fields-based recognition performance.