Skip to Main Content
Table 5 

Comparing our results with some representative state-of-the-art systems.

Bio-NER
Method
F-score
(Okanohara et al. 2006) Semi-Markov CRF + global features 71.5 
(Hsu et al. 2009) CRF + PSA(1) training 69.4 
(Tsuruoka, Tsujii, and Ananiadou 2009) CRF + SGD-L1 training 71.6 
Our Method CRF + ADF training 72.3 
 
Segmentation Method F-score 
(Gao et al. 2007) Semi-Markov CRF 97.2 
(Sun, Zhang, et al. 2009) Latent-variable CRF 97.3 
(Sun 2010) Multiple segmenters + voting 96.9 
Our Method CRF + ADF training 97.5 
 
Chunking Method F-score 
(Kudo and Matsumoto 2001) Combination of multiple SVM 94.2 
(Vishwanathan et al. 2006) CRF + SMD training 93.6 
(Sun et al. 2008) Latent-variable CRF 94.3 
Our Method CRF + ADF training 94.5 
Bio-NER
Method
F-score
(Okanohara et al. 2006) Semi-Markov CRF + global features 71.5 
(Hsu et al. 2009) CRF + PSA(1) training 69.4 
(Tsuruoka, Tsujii, and Ananiadou 2009) CRF + SGD-L1 training 71.6 
Our Method CRF + ADF training 72.3 
 
Segmentation Method F-score 
(Gao et al. 2007) Semi-Markov CRF 97.2 
(Sun, Zhang, et al. 2009) Latent-variable CRF 97.3 
(Sun 2010) Multiple segmenters + voting 96.9 
Our Method CRF + ADF training 97.5 
 
Chunking Method F-score 
(Kudo and Matsumoto 2001) Combination of multiple SVM 94.2 
(Vishwanathan et al. 2006) CRF + SMD training 93.6 
(Sun et al. 2008) Latent-variable CRF 94.3 
Our Method CRF + ADF training 94.5 
Close Modal

or Create an Account

Close Modal
Close Modal