Research Article

ChemTok: A New Rule Based Tokenizer for Chemical Named Entity Recognition

Table 6

NER performance (-score in %) of classifiers using BioCreative data set.

TokenizerClassification algorithm
CRF SVM
DevelopmentTestDevelopmentTest

White space75.3975.4475.6575.67
ChemSpot78.4678.8983.2682.88
tmVar76.1576.5082.2982.27
ChemTok81.7781.8985.1584.94