Two-phase biomedical named entity recognition based on semi-CRFs

As a crucial step for the other tasks, such as human gene/protein normalization, relationship extraction and hypothesis generation, biomedical named entity recognition remains a challenging task. This paper represents a two-phase approach based on semi-CRFs and novel feature sets. Semi-CRFs put the label to a segment not a single word which is more natural than the other machine learning methods. Our approach divides the whole biomedical NER into two sub-tasks: term boundary detection and semantic labeling. At the first phase, term boundary detection sub-task detects the boundary of the entities and classifies the entities into one type C. At the second phase, semantic labeling sub-task label the entities detected at the first phase the correct entity type. To make a comparison, experiments conducted both on CRFs model and semi-CRFs model at each phase. Our experiments carried out on JNLPBA2004 datasets achieve an F-score of 73.20% based on semi-CRFs without deep domain knowledge and post-processing algorithm, which outperforms most of the state-of-the-art systems.