Classifiers ensemble approaches for automatic recognition of biomedical named entities

As a new branch of data mining and knowledge discovery, the research of biomedical text mining has a rapid progress currently. Biomedical named entity recognition is a basic technique in the biomedical knowledge discovery and its performance has direct effects on further discovery and processing in biomedical texts. In this paper, we present classifiers ensemble approaches for biomedical named entity recognition. Four individual classifiers, Generalized Winnow, Conditional Random Fields, Support Vector Machine, and Maximum Entropy are combined through three different strategies. We demonstrate the effectiveness of the strategies and compare their performances with standalone classifier system. The experiments are carried on JNLPBA2004 corpus with an F-sore of 77.57%. Experimental results show that the proposed method, stacking ensemble strategy, can yield promising performances.