Skip to main content

Table 5 Comparison of training time between CLSTM (contextual long short-term memory networks [LSTMs] with conditional random fields [CRF])and comparative methods for the NCBI, GM, and CDR corpora

From: Biomedical named entity recognition using deep neural networks with contextual information

Training time (Hours)NCBIGMCDR
BiLSTM4.0810.774.39
BiLSTM-CRF4.7012.565.23
GRAM-CNN11.2734.0812.64
BERT1.0110.043.72
CLSTMword level5.5414.575.98
 character level4.8413.305.64
 word+char levels5.8414.735.91
 Average5.4114.205.84