Biomedical named entity recognition using conditional random fields and rich feature sets：（生物命名实体识别使用条件随机域和丰富的功能集）.pdfVIP

Appears in Proceedings of the COLING 2004 International Joint Workshop on Natural Language Processing in Biomedicine and its Applications (NLPBA). Geneva, Switzerland. 2004. A demo of the system can be downloaded from /~bsettles/abner/ . Biomedical Named Entity Recognition Using Conditional Random Fields and Rich Feature Sets Burr Settles Department of Computer Sciences Department of Biostatistics and Medical Informatics University of Wisconsin-Madison Madison, WI, USA bsettles@ 1 Introduction particular have been shown to be useful in part- of-speech tagging (Lafferty et al., 2001), shallow As the wealth of biomedical knowledge in the parsing (Sha and Pereira, 2003), and named en- form of literature increases, there is a rising need tity recognition for newswire data (McCallum for effective natural language processing tools and Li, 2003). They have also just recently been to assist in organizing, curating, and retrieving applied to the more limited task of finding gene this information. To that end, named entity and protein mentions (McDonald and Pereira, recognition (the task of identifying words and 2004), with promising early result